ES2427216T3 - Bioavailable curcuminoid formulations for the treatment of Alzheimer's disease and other age-related disorders - Google Patents

Abstract

A curcuminoid composition that offers improved oral bioavailability, comprising: a curcuminoid; an antioxidant; a pharmaceutically acceptable, water-solubilizing vehicle, comprising: (a) lipid micelles where the composition is provided in the form of microemulsion, solid nanoparticles or lipid micelles; or (b) a microencapsulated oil where the composition is provided as a microencapsulated oil; optionally, a glucuronidation inhibitor, and where the composition is suitable for oral administration.

Description

Bioavailable curcuminoid formulations for the treatment of Alzheimer's disease and other age-related disorders

Field of the Invention

In general, the present invention relates to curcuminoid compositions useful for the treatment and prevention of disease, in particular age-related diseases.

Background of the invention

The pathogenesis of diseases such as Alzheimer's and Parkinson's reveals increasing evidence of oxidative damage and inflammatory factors. Unfortunately, despite a solid epidemiology and justification, in general, the approaches with antioxidants and NSAIDs for these diseases associated with age have not prospered in clinical practice. For example, vitamin E has failed in trials on prevention of Alzheimer's disease and heart disease, while cyclooxygenase (COX) inhibitors have failed to treat Alzheimer's disease and its use has been limited to prevention efforts with other traditional antioxidants (selenium, vitamin E, β carotene), estrogens and COX-2 inhibitors. The demography of the aging of our population reveals an urgent need for adequate alternatives for the prevention and possible treatment of one or more of the chronic diseases associated with aging.

In the form of turmeric extract, curcumin is the yellow substance that is used in curry as a food condiment, for example, in yellow mustard. Like aspirin, the "miracle remedy" that remains one of the few prevention agents that have prospered, the short-term potential that curcumin offers for health has a long history and a relatively well established scientific basis. It has been identified as an important bioactive agent in an empirically developed system of traditional Indian and Chinese medicine.

Curcumin (diferulomethane) is not only a potent antioxidant and natural anti-inflammatory agent, which acts on pro-inflammatory mediators regulated by NFkB and AP-1, including COX-2, iNOS, il-1 and TNFα, but also has numerous useful activities and has demonstrated its therapeutic potential in multiple models of preclinical and animal cultures for diseases often related to aging. These diseases include cancer (colon, prostate, breast, skin, leukemia, etc.) (Agarwal et al., 2003), prion disease (Caughey et al., 2003), atherosclerosis (Miquel et al., 2002; Ramaswami et al., 2004), stroke (G. Sun, personal com.), Alcoholic toxicity over the CNS (Rajakrishnan et al., 1999), brain trauma (F. Gomez-Pinilla, personal com. At UCLA), disease of Huntington (MF Chesselet, personal com.), Marie-Charcot Tooth (J. Lupski, personal com.), multiple sclerosis (EAE), and Alzheimer's disease.

Curcumin can block the aggregation of Aβ and other amyloid plaque-forming peptides to toxic oligomers and fibrils, chelate metals that cause oxidative damage of lipids, proteins and DNA in the brain, inhibit aberrant inflammation through transcription of NFkB and AP-1, stimulate beneficial microglial phagocytosis such as amyloid vaccine to eliminate beta-amyloid plaques in the brain, and inhibit the production of BACE under conditions of inflammation and oxidative damage, thus limiting the production of Aβ. The efficacy of broad spectrum for age-related disease is accentuated by positive data in terms of increased life expectancy (Kitani et al., 2004). Based on its outstanding safety and efficacy profile in multiple disease models with oxidative damage and inflammatory factors, curcumin has demonstrated excellent potential for disease pathogenesis in Alzheimer's models. Curcumin is on the reduced list of agents useful for cancer chemoprevention that is being developed by the National Institute Against Cancer (NCI), which studies curcumin through the National Toxicology Program and preclinical efficacy and safety trials (Kelloff et al., 1996; Chainani-Wu, 2003). Curcumin has undergone several phase I clinical trials for cancer and is currently the subject of other clinical trials on cancer in multiple centers in the United States and other countries.

The structure of curcumin resembles that of amyloid-binding compounds. (FIG. 1) It is known that amyloid dyes, such as Congo Red (CR) bind through planar hydrophobic subgroups with a conveniently separated charge and suppress the toxicity and aggregation of β-amyloid and other beta-sheet dependent peptides . The Congo Red (CR) analog, Chrysamine G, is more permeable in the brain and maintains the amyloid binding properties of CR. Curcumin shares the separation of the 19 angstrom CR between its groups of polar phenols, is very permeable in the brain and binds to the amyloid peptides, inhibiting their aggregation and toxicity in vitro. We have discovered that curcumin effectively reduces amyloid accumulation in vivo in APP Tg mice. Since the CR anti-amyloid junction is generic and potentially relevant to other intraneuronal beta-sheet aggregates, including Huntington, a-synuclein, prions and tau, curcumin's anti-amyloid activity may be more relevant. beyond the extracellular amyloid for intraneuronal aggregates. In fact, curcumin is one of the most effective anti-prion compounds that have been tested in vitro, although it did not work in vivo with an oral dose of an undisclosed formulation (Caughey et al., 2003). This

increases the limitations of the oral bioavailability of curcumin, the object of the present invention.

The benefits of curcumin as a treatment for multiple diseases with accumulation of amyloid proteins and other disorders of CAG repetitions are being established, and its effectiveness for the treatment of stroke, head trauma, metabolic syndrome and many other disorders is also beginning to be understood. , which include some forms of cancer and arthritis, as well as to promote wound healing. However, all these therapeutic applications are limited due to poor intestinal absorption.

Although curcumin is an effective medication in multiple animal models for human diseases when administered in food at high doses (typically 2,000-5,000 ppm in the diet in cancer trials), its bioavailability is currently thought to be available It is so poor that it cannot be used for treatment outside the colon in humans. Curcumin is very hydrophobic and usually does not dissolve when administered as a powdered extract in common nutraceuticals. Most curcumin activities require levels of 100-2,000 nanomolars (0.1-2 micromolars) in vitro, although current supplements cause negligible nanomolar plasma concentrations (see Sharma et al., 2004). The group of R. Sharma de Leicester has tried repeatedly and has not achieved significant blood levels, beyond the low nanomolar range (Garcea G., Jones JD, Singh R., Dennison AR, Farmer PB, Sharma RA , Steward WP, Gescher AJ, Berry DP. Detection of curcumin and its metabolites in hepatic tissue and portal blood of patients following oral administration. Br J Cancer. March 8, 2004; 90 (5): 1011-5. PMID: 14997198 .) These, among others, concluded that the administration of effective concentrations of oral curcumin in systemic tissues (outside the GI tract) "is probably not feasible." Most of the literature shares this opinion, which has led the NCI to focus on colon cancer.

Three factors limit the absorption of curcumin and should be addressed: 1) the rapid glucuronidation / sulfation of phenolic hydroxyl groups of curcumin and the high "first-pass" clearance; 2) curcumin is unstable in aqueous solution at pH 7 and higher; and 3) curcumin is very hydrophobic and is not normally soluble in water at an acidic pH or when administered as dry powder in existing supplements. (Most curcumin is never absorbed and simply passes through the GI tract and is excreted.)

Solubilization is critical to avoid this, but curcumin requires a pH 8.5 to dissolve completely. For this reason, cancer patients are given high doses, typically up to 8 grams per day. Diarrhea is a common side effect. Garcea, G. et al. (2004) reports that patients who receive 3.6 grams of curcumin per day (using a standard powder extract capsule supplied by Sabinsa Corporation), reach negligible levels in blood and liver. They conclude that "the results suggest that the doses of curcumin necessary to reach sufficient liver levels to exert pharmacological activity are probably not viable in humans."

Curcumin is not soluble at acidic pH and decomposes when solubilized and dissolved in water at a neutral pH.

or alkaline (for example, in the GI tract, after the small intestine), due to the keto-enol transformations in the β-diketone bridge. On the other hand, curcumin is susceptible to rapid glucuronidation / sulfation. The main supplier of the United States, Sabinsa, has tried to manufacture a more bioavailable form, adding bioperine (in development) to inhibit glucuronidation. However, this approach is wrong, because most of the glucuronidation occurs in the upper GI tract, where the pH is acidic, and curcumin does not dissolve completely until pH 8.5 and higher. Even worse, inhibition of glucuronidation can cause serious health risks. Glucuronidation protects against many toxins and contributes to the metabolism of commonly used drugs. The majority of elderly patients receive multiple drugs, at levels that are likely to be altered in an unsafe way by glucuronidation inhibition.

Although many groups have raised theoretical ideas to improve curcumin absorption, most have been part of completely in vitro studies, probably due to the difficulty involved in measuring curcumin and its metabolites in tissue. Our facilities, knowledge and experience in the measurement of curcumin and tetrahydrocurcumin have allowed a better approach.

Summary of the Invention

In accordance with the claims, the present invention provides curcumin compositions that have improved bioavailability and that can be used as therapeutic agents to treat (and possibly slow down or prevent) a number of diseases and disorders associated with age. In one aspect of the invention, a curcuminoid composition that offers improved bioavailability comprises a curcuminoid, an antioxidant and a pharmaceutically acceptable water-solubilizing vehicle and, optionally, a glucuronidation inhibitor. In one embodiment, solubilization is achieved by forming curcuminoidelipid micelles and the composition is provided in the form of a microemulsion or solid lipid nanoparticles (SLN). In another embodiment, the curcuminoid is dissolved in an edible oil, which can then be microencapsulated or emulsified. The composition can be provided in the form of gel, capsule, liquid and other pharmaceutically acceptable form.

In accordance with the claims, the invention also provides curcuminoid compositions for use in a method of treating, slowing down and / or preventing disease, particularly of neurodegenerative diseases associated with age, such as Alzheimer's, and comprises the administration of a Therapeutically effective dose of a solubilized curcuminoid, resistant to hydrolysis and, optionally, resistant to glucuronidation, according to the claims. Functional doses of curcuminoids can be administered to mammals, such as rats, mice and humans, using an improved curcuminoid composition, as described herein.

Brief description of the illustrations

These and other features and advantages of the invention will be better understood when considered in conjunction with the following detailed description and by consulting the accompanying illustrations, where:

FIG. 1 illustrates the molecular structures and similarity of the molecular separation of Congo Red, curcumin and Chrysamine G;

FIG. 2 is a graph of the absorbance versus time for the antioxidant protection activity (stabilization) of various formulations of curcumin, tetrahydrocurcumin and ascorbate;

FIG. 3 is a graph of ascorbate concentration versus time for various curcumin and ascorbate formulations, at 450 nm;

FIG. 4 is a graph similar to that of FIG. 3, but with the absorbance measured at 405 nm.

Detailed description of the invention

According to the claims, the present invention provides curcuminoid compositions that have improved bioavailability. In one aspect, this composition comprises a curcuminoid, an antioxidant (which stabilizes the curcuminoid against hydrolysis), a water solubilizer, a pharmaceutically acceptable carrier and, ideally, a glucuronidation inhibitor. In some embodiments, for example when the curcuminoid is tetrahydrocurcumin, which is stable even at an alkaline pH, it can be dispensed with the antioxidant. In various disclosures, the composition is provided in the form of microemulsion, solid lipid nanoparticles (SLN), microencapsulated oil, gel, capsule, liquid and / or other pharmaceutically acceptable form, and is suitable for administration to a human or other mammal by enteric, parenteral, topical or by some other mode of administration.

Some examples of curcuminoids, by way of example only, include curcumin, tetrahydrocurcumin, demethoxycurcumin, bisdemethoxycurcumin, curcumin esters (which function as prodrugs) and combinations thereof. The combination of curcumin and its metabolite, tetrahydrocurcumin, is particularly effective. Tetrahydrocurcumin is relatively stable in water at a physiological pH, is a competitive inhibitor of glucuronidation, improves absorption and, ultimately, plasma curcumin levels.

To protect curcuminoid from hydrolysis, a water soluble antioxidant is used. Some examples, by way of example only, include ascorbic acid (ascorbate, vitamin C and its water-soluble acylated derivatives), α-lipoic acid (alpha-lipoate), vitamin E and derivatives, N-acetylcysteine (NAC), and glutathione reduced (GSH). Even tetrahydrocurcumin provides curcumin with some protection against hydrolysis. Mixtures of antioxidants can also be used.

Most of the previous efforts to explain the poor bioavailability of curcumin have focused on rapid glucuronidation. However, tetrahydrocurcumin and curcumin are similarly hydrophobic and poorly soluble in water, in addition to undergoing glucuronidation and sulfation equally rapidly. However, our data and bibliography data show that tetrahydrocurcumin has a much greater bioavailability than curcumin, with plasma levels derived from the same oral doses between 7 and 8 times higher. The enhanced bioavailability of tetrahydrocurcumin is largely attributed to its being stable even at a basic pH. On the contrary, curcumin is unstable in aqueous solutions with a pH greater than 7.0 and, therefore, is unstable in the intestinal tract, where absorption mainly occurs. It is hydrolyzed to degradation products of ferulic acid and vanillin. It is expected that a preparation that stabilizes curcumin at a slightly more basic pH will increase the bioavailability of solubilized curcumin up to 7 or 8 times. The combination of curcumin with an additional water-soluble antioxidant, such as ascorbate, can stabilize curcumin at a pH of 7.4.

We have tested serial dilutions of ascorbate and think that stabilization is effective up to 4 hours (a typical absorption cycle) with curcumin / ascorbate ratios of up to 16: 1, resulting in a slight reduction in efficacy between 8: 1 and 16 :one. Therefore, the supply of a curcuminoid / antioxidant ratio of approximately 10: 1

or higher would be enough to avoid hydrolysis during absorption and improve bioavailability. As an example, purely enunciative, a preparation containing 330 mg of curcumin (MW 368) would have only an additional 17.75 mg of ascorbate (MW 198). Therefore, it is not necessary to add a prohibitive amount of antioxidant to the whole of a formulation.

Some examples of glucuronidation inhibitors include tetrahydrocurcumin, piperine (Bioperine®),

5 probenecid (Probene®), and diclofenac. (See, for example, Uchaipichat V, Mackenzie PI, Guo XH, Gardner-Stephen D, Galetin A, Houston JB, Miners JO. Human udp-glucuronosyltransferases: isoform selectivity and kinetics of 4-methylumbelliferone and 1-naphthol glucuronidation, effects of organic solvents, and inhibition by diclofenac and probenecid. DrugMetab Dispos. 2004 Apr; 32 (4): 413-23; Vree TB, van den Biggelaar-Martea M, Verwey-van Wissen CP, van Ewijk-Beneken Kolmer EW. Probenecid inhibits the glucuronidation of indomethacin and O

10 desmethylindomethacin in humans. A pilot experiment Pharm World Sci. 1994 Feb 18; 16 (1): 22-6.) Mixtures of inhibitors can also be used. Curcumin and tetrahydrocurcumin are rapidly eliminated by first-pass glucuronidation and it has been documented that glucuronidation with piperine improves the bioavailability of curcumin in humans. However, the addition of a glucuronidation inhibitor poses problems of interactions with many other drugs that are eliminated through glucuronidation. The present invention

15 may improve bioavailability with this inhibitor, although it is included as an optional component in some formulations.

Despite its reduced bioavailability, oral administration of curcumin in vivo can competitively and temporarily inhibit glucuronidation of other drugs, such as mycophenolic acid. (Basu et al Nikhil K. Basu, 20 Labanyamoy Kole, Shigeki Kubota, and Ida S..Owens, Human UDP-Glucuronosyltransferases Show Atypical Metabolism of Mycophenolic Acid and Inhibition by Curcumin. Drug Metabolism and Disposition 32: 768-773, 2004) . Tetrahydrocurcumin, which offers levels between 7 and 8 times higher, is expected to offer much greater bioavailability and stability and, therefore, will be more effective than curcumin. On the other hand, it is anticipated that tetrahydrocurcumin will protect curcumin against glucuronidation. Likewise, data from our in vitro bioassay show that, at a rate of 1: 1, tetrahydrocurcumin and curcumin establish synergies together in vitro. Therefore, the formulation of tetrahydrocurcumin with curcumin at a molar ratios of 1: 1

or higher will improve the bioavailability of curcumin and provide efficacy associated with synergy.

Glucuronidation and sulfation of curcuminoids can also be inhibited by esterifying one fatty acid or another

30 acyl group with one or both hydroxyls in the two methoxyphenol groups that are the targets of rapid enzymatic glucuronidation and sulphation. The blocking ester groups will then be removed from the curcuminoid prodrug ester in vivo by esterases in the target tissues, providing free curcumin. In principle, any fatty acid can be used for esterification. Since our data demonstrate effective synergies in vivo between omega-3 fatty acid and docosahexaenoic acid (DHA) and curcumin in an Alzheimer's model, the

35 DHA is an attractive option. DHA is not only selectively assimilated by the brain, where curcumin is neuroprotective in multiple systems, but it can also direct drugs to tumor cells, a clinically important target of curcumin. (For the use of DHA directed at lipophilic cancer drugs against tumor cells, see Bradley MO, Swindell CS, Anthony FH, Witman PA, Devanesan P, Webb NL, Baker SD, Wolff AC, Donehower RC, Tumor targeting by conjugation of DHA to paclitaxel. J Control

40 Release July 6, 2001; 74 (1-3): 233-6; Harries M, O'Donnell A, Scurr M, Reade S, Cole C, Judson I, Greystoke A, Twelves C, Kaye S; Phase I / II study of DHA-paclitaxel in combination with carboplatin in patients with advanced malignant solid tumors. Br J Cancer. November 1, 2004; 91 (9): 1651-5.) Also, long-chain polyunsaturated acid acids such as DHA favor the administration of the drug in the lymphatics, which reduces the problems of first-pass metabolism. On the other hand, chain polyunsaturated fatty acids

Long, like DHA, increases the administration of the drug in the lymphatic system, which reduces the problems of first-pass metabolism.

Thus, it is possible to synthesize a "prodrug" of DHA-curcuminoid from a curcuminoid (such as curcumin) and a DHA, joining the DHA to curcumin in a phenolic hydroxyl with a stoichiometry of 1: 1, 50 analogous to used by Bradley et al 2001 for paclitaxel. (Bradley MO, Webb NL, Anthony FH, Devanesan P, Witman PA, Hemamalini S, Chander MC, Baker SD, He L, Horwitz SB, Swindell CS. Tumor targeting by covalent conjugation of a natural fatty acid to paclitaxel. Clin Cancer Res October 2001; 7 (10): 3229-38.) Alternatively, DHA can be linked to curcumin through a multi-step program, analogous to that used to bind doxorubicin to curcumin (Wang Y, Li L, Jiang W, Yang Z, Zhang Z. Synthesis and preliminary antitumor activity

55 evaluation of a DHA and doxorubicin conjugate. Bioorg Med Chem Lett. June 1, 2006; 16 (11): 2974-7.) DHA-curcumin esters will provide a neuroprotective DHA and a neuroprotective curcumin with superior shelf life and improved bioavailability.

More generally, a curcuminoid ester, in accordance with one aspect of the invention, may comprise the

The esterification product of a curcuminoid and a carboxylic acid RCOOH, where R is selected so that the resulting ester is not toxic and is subject to cleavage by an in vivo esterase. A preferable class of carboxylic acids for this purpose are fatty acids, especially essential fatty acids, such as DHA (an omega-3 fatty acid). Monoglycerides, diglycerides and triglycerides are other examples of carboxylic acids suitable for the production of curcuminoid prodrugs.

65 Curcuminoid is solubilized using any number of water solubilizing vehicles. For the purposes of the present, "water solubilizing vehicle" refers to an agent, composition, compound or medium that provides a curcuminoid in a more water soluble or water dispersible form, or that interacts with the curcuminoid

5 to print greater dispersibility or water solubility. In general terms, two examples of vehicle categories include lipid micelles and microencapsulated oils.

Lipid micelles containing a curcuminoid can be manufactured with any of a variety of lipids. Some examples, by way of example only, include (i) fatty acids, such as stearic acid; (ii) phospholipids, for example, phosphoglycerides, such as phosphatidylcholine ("PC"); phosphatidylethanolamine, phosphatidylinositol; (iii) bile acids, for example, deoxycholic acid (deoxycholate) and conjugates thereof (for example, amino acid conjugates, such as glycocholate and taurocholate); (iv) edible oils, especially healthy oils, for example, vegetable oils, olive oil, rapeseed oil, fish oil; (v) triglycerols; (vi) mixtures of any of these and / or other lipids and derivatives, for example, pharmaceutically acceptable salts, hydrates and

15 conjugates thereof. The combination of a phospholipid (for example, PC or soy lecithin or egg lecithin) and another surfactant, such as bile acid / salt (for example, deoxycholate, taurocholate); propylene oxide / ethylene oxide copolymers (Poloxamer 188, Poloxamer 182, Poloxamer 407, Poloxamine 908), or propylene oxide / ethylene oxide sorbitan copolymers (Polysorbate 20, Polysorbate a60, Polysorbate 80)) is particularly useful. Other useful lipids include natural lecithin (a mixture of glycolipids, triglycerides and phospholipids, including PC). Generally, long chain compounds are preferred over short chain compounds when the composition is provided in the form of an emulsion.

In one embodiment, micelles are prepared using a modified protocol from Began et al 1999. (Began G, Sudharshan E, TJdaya Sankar K, Appu Rao AG. Interaction of curcumin with phosphatidylcholine: A

25 spectrofluorometric study. J Agric Food Chem., December 1999; 47 (12): 4992-7.) Thus, PC-curcumin micelles can be prepared using a mixture of phosphatidylcholine (PC) and deoxycholate (DOC), at a DOC / PC molar ratio of 2.0. Curcumin (10 mg) is added to the DOC / PC mixture, solubilized together in a chloroform / methanol mixture (2: 1 by volume). Subsequently, the mixture is evaporated and dried with nitrogen. The resulting thin film is resuspended in 2 ml of phosphate buffer solution (PBS), at a pH of 7.4 and then the solution (5 mg / ml) is subjected to ultrasound for 5 minutes using an ultrasonic bath.

Began et al performed all the experiments in vitro using a 2: 1 molar ratio of DOC / PC, and found that the micelles were saturated with curcumin at a molar ratio of 6: 1 or 7: 1 of PC / curcumin. In contrast, in one embodiment of the invention, micelles are formed with a DOC / PC molar ratio of 3.7: 1 and a molar ratio of

35 pcs / curcumin of 1: 3.33 (see Example 1). In general, surfactant / lipid ratios (eg DOC / PC ratios) of 1: 1 to 4: 1 are used. Theoretically, bioavailability can be further improved by increasing the PC / curcumin ratio, to a saturation ratio of 6: 1, but the PC volume will reduce the amount of curcumin that can be obtained per 1 gram capsule.

In some embodiments, the micelles also contain one of the antioxidants, or both, and the glucuronidation inhibitor. Alternatively, the last two components are added to the composition after micelle formation. In either case, the micelles can be administered in various clinically viable forms (for example, diluted in saline and administered in liquid form; conjugated with a binder and other excipients, and administered in capsule form, etc.).

It is also possible to form lipid micelles in vivo, for example by dissolving a curcuminoid in an edible oil, adding an antioxidant and a glucuronidation inhibitor; microencapsulating the combined components and allowing bile acids to form micelles in vivo.

The second broad category of "water-soluble vehicles" are microencapsulated oils, in which the micro-drops of a non-toxic oil, preferably an edible oil, especially healthy oils (for example, vegetable oils, olive oil, rapeseed oil and fish oil) are surrounded by a coating, blister or other membrane that is water soluble or decomposes in water. Microencapsulated oils are known in the pharmaceutical field and can be prepared by various methods, including coating.

55 by air suspension, centrifugal extrusion, vibration nozzle and spray drying. In one embodiment, the coating dissolves or decomposes in the GI tract with zero order kinetics or first order kinetics. Microencapsulated oils may be administered orally (for example, with the desired formulation to provide a capsule or other administration system) or by another appropriate route of administration. The particles can be stabilized with polysorbate 80, for example, standard gel capsules or chitosan coated gel capsules can also be used to contain the formulations and promote intestinal administration.

In another embodiment of the invention, the composition is provided in the form of solid lipid nanoparticles (SLN), which can be administered by various clinical forms, for example in gel capsules, or by formulation with one or more binders or other excipients. . SLN preparation is known in the field. See, for example, Luo, Y., Ren, L., Zhao, X., Qin, J., Solid Lipid Nanoparticles For Enhancing Vinpocetine's Oral

Bioavailability, J. Controlled Release, 114 (2006) 53-59 (available online, at www.sciencedirect.com). Some illustrative examples of lipids suitable for SLN include: triacylglycerols, such as tricaprine, trilaurine, trimiristine, tripalmitin, tristearin; acylglycerols, such as glycerol monostearate, glycerol behenate, glycerol palmitoestearate; fatty acids, such as stearic acid, palmitic acid, decanoic acid, behenic acid; waxes, such as cetyl palmitate; cyclic complexes, such as cyclodextrin, para-acyl-calix-arenes; and combinations of the same.

Some examples of suitable surfactants include the following: phospholipids, such as soy lecithin, egg lecithin, phosphatidylcholine; propylene oxide / ethylene oxide copolymers, such as Poloxamer 188, Poloxamer 182, Poloxamer 407, Poloxamine 908; propylene oxide / sorbitan ethylene oxide copolymers, such as Polysorbate 20, Polysorbate 60, Polysorbate 80; polymers of alkylaryl polyether alcohol, such as tyloxapol.

Some examples of suitable co-surfactants include the following: sodium cholate, sodium glycocholate, sodium taurocholate, sodium taurodeoxycholate.

There are multiple ways to prepare SLN formulations. By way of general example only, heat homogenization can be used, which consists of heating a lipid at approximately 10 degrees above its melting temperature, adding a lipophilic drug and dissolving it in the lipid, adding a surfactant to a surfactant ratio. / lipid 1: 1 to 2: 1, cool the combination and then disperse it by ultrasound (typical particle size: ~ 100 mm). This method can be reproduced on a larger scale with ease. (However, evaporation methods are more difficult to reproduce on a larger scale.) Generally, in the case of surfactants such as lecithin, deoxycholate, taurocholate or commercial surfactants such as Tureen 80 and polyoxyethylene hydrogenated castor oil (Cremaphor) , increasing the ratio of surfactant / lipid to 1.5: 1.75 or even 2: 1 tends to reduce particle size and improve absorption. Natural surfactants are preferable, at least in part, because curcumin is normally marketed as nutraceutical and there is a perception that "the natural" is safer.

Effectiveness. In vivo, data acquired from administration to mice of freshly dissolved curcumin on a base (0.5M NaOH) and rapidly diluted in a neutralizing PBS (where it remains stable), administered by probe or via i.p. or im, demonstrated that plasma levels in the micromolar range of 0.25-0.5 provided sufficient brain levels to reach IC50 for a 50% inhibition of proinflammatory IL1β, iNOS and a MAP kinase associated with death cellular (active c-Jun kinase (p) JNK).

Establishing a comparison, based on drug weight / body weight, which generally exaggerates doses in humans, this means 1 mg / 30 g = 33.3 mg / kg and translates approximately to 2.5 g for a person of 75 kg, which is within the range administered to patients in clinical trials.

Thus, an example of a practical formulation that combines the three principles (lipid solubilization, antioxidant protection and competitive inhibition and synergy with tetrahydrocurcumin) for humans would be 165 mg curcumin, 165 mg tetrahydrocurcumin, 17.7 mg stabilizing antioxidant (ascorbate, lipoate, NAC, GSH, etc.), and the rest of DOC / PC micelles as mentioned above or in the form of other amphiphilic lipids, of a 1 g capsule. Binders, processing aids and other excipients may also be present.

Next, selected curcuminoid formulations are collected which can be prepared in accordance with the present invention.

Formula 1. Curcumin / tetrahydrocurcumin micelles with PC-DHA

Curcumin, tetrahydrocurcumin (promotes absorption, establishes synergies with curcumin), phosphatidylcholine (emulsifies, forms micelles with deoxycholate salts, notably promotes absorption), docosahexaenoic acid (from marine oil, establishes synergies with curcumin, facilitates the administration of curcumin in the brain or tumors), an antioxidant or combination of antioxidants (vitamin C or lipidated vitamin C, alpha-lipoic acid, vitamin E) (used to stabilize and recycle curcumin, maintaining its stability and preventing it from becoming a prooxidant in the capsule or in the body).

Formula 2. Curcumin Micella with PC-DHA

Curcumin (we have determined that 25% of plasma curcuminoids naturally convert to tetrahydrocurcumin, which favors absorption and establishes synergies with curcumin), phosphatidylcholine, docosahexaenoic acid, an antioxidant or combination of antioxidants (vitamin C or lipidated vitamin C , alfalipoic acid, vitamin E).

Formula 3. Curcumin with DHA / olive oil

Curcumin, docosahexaenoic acid, an antioxidant or combination of antioxidants (vitamin C or vitamin C

lipidated, alpha-lipoic acid, vitamin E). Our data reveal that when curcumin dissolves in oil, plasma curcumin remains at low levels, although curcumin in red blood cells remains quite high, which explains bioavailability despite insignificant plasma levels. No other group has documented this fundamental observation, which seems to be the simplest method of promoting bioavailability. Other healthy oils (fish oil, rapeseed oil, other oils with a high omega-3 content) can be used. In this formulation, the curcumin dissolved in oil can be microencapsulated.

Formula 4. Curcumin-DHA ester

Prodrug A new compound that solves the problems of curcumin bioavailability and efficacy in humans, which offers optimal oral absorption, antioxidant activities in the brain, antiamyloidogenic properties (for example, DHA reduces the production and accumulation of amyloid), and glucuronidation resistance. The composition may be in the form of a micelle, SLN or other formulation.

Formula 5. Curcumin-DHA ester and tetrahydrocurcumin-DHA ester

Another prodrug with improved oral stability and availability.

Formula 6. Curcumin-SLN

Using fat with a high melting point, solubilized and hydrolysis resistant curcuminoid formulations can be produced in the form of clinically viable solid lipid nanoparticles (SLN), with an internal phase lipid, a drug (curcuminoid), a surfactant and a co-surfactant

In addition to the compositions provided herein, the invention also provides a method for treating various diseases, particularly neurodegenerative, age-associated diseases, such as Alzheimer's disease. In one embodiment, the method comprises administration to a mouse, a rat, a human being or another mammal of an effective dose of a curcuminoid, where the curcuminoid is provided in the form of a composition described herein, for example, in the form of micelles. lipids, microencapsulated oils, SLN, etc., and optionally in the form of gel, capsule or liquid. For humans, most curcumin activities require levels of 100-2,000 nanomolars (0.1-2 micromolars).

The inclusion of a solubilizing lipid greatly increases the levels of curcumin in plasma and red blood cells. In one experiment, DHA / lecithin micelles were prepared in a ratio of 1: 1, placing the curcumin in hot DHA (55 ° C) and sonicating it to disperse the curcumin provided by the PC-DHA micelles (but without an antioxidant or an independent glucuronidation inhibitor), and a curcumin concentration of 0.29 mg / ml in plasma and 0.96 in red blood cells (0.8 and 2.6 mm, respectively) was achieved. On the contrary, using the Sabinsa formulation (with piperine, but without a water-solubilizing vehicle) a curcumin concentration of 0.17 mg / ml in plasma (0.46 mm) was obtained. Long-chain fatty acids, particularly unsaturated fatty acids such as DHA, have added value with regard to the direction of lipophilic drugs to the lymphatic system and the reduction of high losses in first-pass metabolism.

Examples

Illustrative examples of the invention, its preparation and its use in vivo are included below.

materials

Stearic acid, L-phosphatidylcholine (lecithin), sodium taurocholate, deoxycholate and doubly distilled nanopure water (dd) are from Sigma. DHA and curcumin (97%) are from Cayman Quemicals (Michigan, MI). Tetrahydrocurcumin (99%) is from Sabinsa Corporation (Piscataway, NJ).

General protocol for nanoparticles - Heat homogenization (modified from Bocca et al., 1998)

one.

 Prepare a water bath at about 75-80 ° C to melt the lipid.

2.

Measure lipids, surfactants and water for each test formula and batch size.

3.

Add the internal phase lipid (stearic acid or DHA) and the drug and lecithin (surfactant) with magnetic stirrer to empty the coated vessel.

Four.

Add sodium taurocholate (co-surfactant) to the water and continue stirring.

5.

Heat the water solution dd at 75 ° C on a heated plate.

6.

Once the lipid has melted, add the water solution to the coated container.

7.

Allow the combined lipid / water solution to equilibrate at 75 ° C.

8.

Disperse the solution in an Ika Ultra-Turrax T 18 rotor-stator homogenizer (with 19mm rotor-stator) at 30,000 rpm for 2 minutes.

9.

Quickly inject / disperse 1 ml of tempered microemulsion through a small gauge needle

in 10-20 ml of water dd at 2 ° C in a glass vial with constant agitation.

10.

Wash three times in water dd by diafiltration, using a TCF2 device with a YM100kD cutting membrane (Amicon, Danvers, Ma).

eleven.

Store the product of the lipid nanoparticle at 2 ° C.

5 12. Measure the particle size.

Optimization In order to achieve a target of 100 nm in size for parenteral administration and a polydispersity target of 0.10, stearic acid has generally been selected as the lipid to be used. However, larger particles may be acceptable for oral administration. Absolute quantities depend on batch sizes, although appropriate stoichiometry is known in the literature. Using this protocol, it has been found that a ratio of lecithin (phosphatidylcholine) / sodium taurocholate of 2: 3, a surfactant / lipid ratio of 1: 4 and a mixing time of 70 seconds with a mixing speed of 18,000 rpm they are optimal for beta-carotene, which provided an average entrapment efficiency of 40% and a concentration of 0.22 mg / ml (Triplett, 2004).

15 Preparation of solid lipid nanoparticles (SLN)

Starting formula. 0.710 molar fraction of stearic acid; 0.210 molar fraction of lecithin; 0.069 taurocholate molar fraction; Curcumin or other curcuminoid gradually varies around the 0.011 molar fraction. The stearic acid lipid is maintained at approximately 75 ° C to completely melt it. Separately, double distilled water is heated to 75 ° C. Typically, the surfactants are added to the water under magnetic stirring and allowed to equilibrate at 75 ° C. The water-surfactant solution is added to the molten lipid and allowed to equilibrate at 75 ° C. Subsequently, the Ika Ultra-Turrax T 18 rotor-stator homogenizer is used to achieve adequate mixing, typically 18,000-30,000 rpm for 70-150 sec. Once mixed, the dispersed lipid phase of

The emulsion solidifies, in order to produce the solid lipid nanoparticles by dispensing through a small gauge needle aliquots of the 1 ml emulsion in almost cold cold water (about 2 ° C) with constant agitation, at a ratio of 1: 20 (tempered microemulsion: cold water). The final product is washed three times with distilled water and subjected to filter sterilization with an Amicon Diaflo apparatus with YM100 membranes (100,000 Dalton cut). Subsequently it is stored under sterile conditions at 4 ° C until administration by probe. Multiple samples of lipid nanoparticles can be prepared from a batch of microemulsion.

Determine the effectiveness of the incorporated drug: To determine the amount of incorporated drug, the SLN dispersions (or micelles) are freeze dried using a freeze dryer. The SLN without charge of each formulation, that is the SLN that does not carry any drug, is also prepared as described

35 above. The amount of curcumin incorporated into the SLN can then be determined by dissolving the freeze-dried SLN in a solution of 95% ethyl acetate / 5% methanol, evaporating it under N2 gas and resuspending the sample by a C-phase in mobile phase (41 % acetonitrile, 35% deionized water, 23% methanol and 1% acetic acid; v / v / v / v). The sample is analyzed by isocratic HPLC methods with a reverse phase C18 column (3.9 x 150 mm, 5-E m particle size; Waters Corporation, Massachusetts, MA), using ultraviolet (UV) light detection at a wavelength of 262 nm (Health et al., 2003; Health et al., 2005).

Drug loading and release measurements. Ultraviolet-visible photo-spectroscopy (UV-Vis) is used to measure the drug load of lyophilized samples, using ethanol as a solvent to dissolve micelles or

45 dried lipid nanoparticles and eliminate secondary absorption and dispersion errors. All absorption readings are performed at a wavelength at which curcumin (262 nm-UV 426 nm-visible) or a curcuminoid (XnM) has a higher absorption compared to other molecules that absorb light in these studies. Other molecules with a detectable absorbance include sodium taurocholate and lecithin, which can be corrected to assume a linear relationship between absorbance and concentration, and with a calibration curve for each absorbent molecule. These are provided by micelle or SLN controls without a "blank" drug. Given the known concentrations of all components except curcumin or derivatives, the drug concentration can be calculated from the corrected absorbance. Stability is evaluated with different storage methods and at different times of micelle preparation.

55 Characterization of the diameters of the nanoparticles.

Atomic force microscopy. Atomic force microscopy (AFM) can be used to obtain images of the preparation sizes of lipid nanoparticles. Small amounts of the lipid nanoparticle suspension are placed in the substrate and allowed to air dry before capturing the images, noting that this can flatten the particles during drying, resulting in a particle size slightly larger than that obtained with methods of dynamic light scattering.

Example 1. DOC / PC / curcumin micelles and administration by probe (Gav) to mice. To produce 5 mg of curcumin micelles with a DOC / PC molar ratio of 3.7: 1 and a PC / curcumin molar ratio of 1: 3.33, for oral administration to 5 mice at 1 mg per mouse of approximately 30 g, we mix 12,438 mg of DOC with 6,219 mg of PC (2: 1 mass ratio) and 1 mg of curcumin powder. This was dissolved by vortex stirring of

a 2 ml mixture of chloroform / methane (2: 1). The solvent was evaporated and dried under nitrogen. The resulting emulsion film was scraped over 2 ml of phosphate buffer solution (PBS) of pH 7.4 and subjected to ultrasound in an ultrasonic bath for 5 minutes. 1 mg (200 μL) was administered per probe to each mouse to produce a plasma concentration of 0.465 micromolars, representing more than twice the plasma level without the

5 micelles The dry mass of the preparation was 2,865 mg / mg of curcumin or curcuminoids, which would provide about 350 mg of curcumin per 1g capsule or 175 mg per 500mg capsule, in practical amounts.

Example 2. DHA-curcuminoid ester

A curcuminoid ester can be synthesized from a curcuminoid and a DHA, linking DHA to curcumin in a phenolic hydroxyl with a stoichiometry of 1: 1, analogous to that used by Bradley et al 2001 for paclitaxel. To a solution of curcumin (366 mg; 1 mol) in methylene chloride (61 ml) subjected to argon, 4-dimethylaminopyridine (122 mg; 1 mol), 1,3-dicyclohexylcarbodiimide (412.36 mg; 2 mol ), and DHA (329.4 mg; 1 mol). The

The reaction mixture is stirred at room temperature for 2 hours. After dilution with diethyl ether, the reaction mixture is washed with 5% HCl, water and saturated aqueous NaCl. The mixture is dried with sodium sulfate, concentrated and purified by chromatography. The DHA-curcumin conjugate is then used as curcumin would be used, for example in gel formulations or prepared micelles. (See Bradley MO, Webb NL, Anthony FH, Devanesan P, Witman PA, Hemamalini S, Chander MC, Baker SD, He L, Horwitz SB, Swindell CS. Tumor targeting by covalent conjugation of a natural fatty acid to paclitaxel. Clin Cancer Res. October 7, 2001; 7 (10): 3229-38). The ester can be solubilized and combined with an antioxidant and a glucuronidation inhibitor, and formulated as an emulsion, SLN or microencapsulated oil.

Example 3. Antioxidant data

25 To test the effect of an antioxidant on the stability of curcumin, several 25 micromolar curcumin solutions were prepared, in the presence / absence of an antioxidant (ascorbic acid or tetrahydrocurcumin). Using absorbance spectroscopy, the absorbance intensity of curcumin at 426 nm was measured. The results are presented in FIG. 2. "PBS" means phosphate buffer solution (pH 7.4); "Curc" means curcumin; "TC" means tetrahydrocurcumin; "Veh" means vehicle. The data shows that ascorbic acid is very effective by substantially retarding the hydrolysis of curcumin. More data on antioxidants are collected in FIG. 3 and 4, which show the absorbance data of the microplate reader at 450 and 405 nm, respectively (flanking the absorbance peak of curcumin at 426 nm). The data shows that ascorbic acid is very effective in slowing the hydrolysis of curcumin, at a curcumin / ascorbic acid ratio of 1: 1 or

35 even 2: 1, with a slightly reduced effectiveness at higher rates.

The Appendix provides additional data on the differential effects of curcumin and its metabolite, tetrahydrocurcumin.

The invention has been described with reference to various embodiments and examples, although it is not limited to these. Variations can be made without departing from the scope of the invention, which is limited only by the appended and equivalent claims thereof.

APPENDIX

Journal of Biological Chemistry

Differential Effects of curcumin and its metabolite, tetrahydrocurcumin, on the pathogenesis of NEUROINFLAMMATION AND PLATE (DIFFERENTIAL AND ITS EFFECTS OF CURCUMIN Metabolite tetrahydrocurcumin neuroinflammation AND PLAQUE ON PATHOGENESIS) Aynun N. Begum1,3, Mychica R. Simmons-Jones1, 3, Giselle P. Lim1,3, Takashi Morihara1,3,5 Peter Kim1,3, Dennis D. Heath4, Cheryl L. Rock4, Mila A. Pruitt4, Fusheng Yang1,3, Bruce Teter1,3, Marni E. Harris- White1,3, Greg M. Cole1,2,3, and Sally A. Frautschy1,2,3 *

55 Departments of 1Medicine and 2Neurology, UCLA, Los Angeles, CA 90095, and 3Greater Los Angeles Healthcare System, GRECC, Sepulveda, CA 91343, 4Cancer Prevention and Control Program, Moores UCSD Cancer Center, UCSD, La Jolla, CA 92093- 0901, current address Department of Post-Genomics and Diseases, Division of Psychiatry and Behavioral Proteomics, Osaka University Graduate School of Medicine D3, Yamada-oka 2-2, Suita-shi, Osaka 565-0871, JAPAN

Short title: Curcumin and tetrahydrocurcumin suppress neuroinflammation and Aβ (Curc and TC suppress neuroinflammation and Aβ)

Send correspondence to: Sally A. Frautschy, Greater Los Angeles Healthcare System, Veterans Administration

65 Medical Center, Research 151, 16111 Plummer Street, Sepulveda, CA 91343, Tel .: 818-895-5892; Fax: 818-8955835; E-mail: frautsch@ucla.edu

SUMMARY

Curcumin (Curc) is effective in multiple inflammatory and neurodegenerative models. However, its limited

5 bioavailability raises concerns about its use and whether Curc metabolites, such as tetrahydrocurcumin (CT), mediate efficacy. We examined the antioxidant, anti-inflammatory or antiamylogenic effects of Curc and CT by chronic administration with food to elderly Tg2576 APPsw mice or by acute administration to wild mice by probe (gav), intraperitoneal (ip) or intramuscular (im) injections ) in an inflammatory model of lipopolysaccharide injection (LPS, ip). Although the CT scan was more stable and absorbed better, reaching plasma levels between 4 and 7 times higher than the Curc; the brain CT levels were not different from those achieved with the Curc, which suggests a lower bioavailability in the brain of the CT compared to the Curc. Brain levels of CT and Curc not only correlated well with efficacy, but also both compounds reduced inducible nitric oxide synthase (iNOS) stimulated by LPS, nitrotyrosine, 8-iso-prostaglandin (PG) F {super2} { alpha} and

15 carbonyls Curc seemed more effective, since Curc levels were associated with greater inhibition compared to comparable levels of CT. Although CT and Curc were equally potent in reducing interleukin-1 {beta} (IL-1 {beta}] caused by both acute and chronic inflammation (Tg2576), Curc was more effective in reducing the burden of amyloid plaque, insoluble Aβ and carbonyls Surprisingly, CT more effectively reduced soluble A {beta} and the active form of cJun N-terminal kinase (pJNK) .The plasma of the Curc-fed mice showed 25% of free CT With respect to each of them individually, the union of CT and Curc more effectively reduced the toxicity and iNOS caused by A {beta} oligomers, as well as by the aggregation of A {oligomers beta} In summary, the exceptional bioabsorption of CT and its stability in plasma does not imply a better bioavailability in the brain CT and Curc affect the pathogenesis of disease in a different way

25 Alzheimer's (AD), but its synergy offers a justification for its combined use.

INTRODUCTION

Neuroinflammation is a characteristic of many neurodegenerative diseases and is implicated in its pathogenesis. Inflammatory mechanisms that contribute to pathogenesis include the secretion of inflammatory mediators, such as interleukin-1β (1L-1β) (1,2), 1L-1α (3), as well as the generation of reactive oxygen species (ROS ) (4-7) and lipid peroxidation products (such as 8-iso-PGF2α) (8,9). ROS not only damages proteins and lipids in AD, but also causes the activation of JNK (10), a signal transduction event involved in neurodegeneration associated with AD (11, 12) and models transgenic

35 (13). Neuroinflammation can also stimulate iNOS, which catalyzes the production of nitric oxide (NO), a reactive free radical that converts to peroxynitrite (OONO a reaction product of NO. And O2-.) And other species of reactive nitric oxide (RNS) derived from NO (14). RNSs are involved in numerous neuropathological disorders, where they contribute to neuroinflammation and oxidative damage (15, 16). The potential pathogenic role of RNS in Alzheimer's disease is supported by the increased expression of nitric oxide inducible synthase (iNOS) in ovule-containing neurons in AD patients (17) and in microglia and astrocytes surrounding beta amyloid plaques. An increase in RNS occurs in brain neurons with AD (16, 18) and also in glia stimulated by β-amyloid (19). Compounds that target neuroinflammatory mediators, such as ROS, RNS, JNK and inflammatory cytokines involved in pathogenesis, are strong candidates for therapeutic intervention. One of these

45 compounds is Curc, the yellow pigment of the turmeric, a widely used product. The turmeric is on the FDA's list of GRAS (substances recognized as generally safe) and the Curc is antagonistic to numerous steps of the relevant inflammatory cascades of the disease, including transcription of AP-1, activation of NF-KB , iNOS and JNK (20-25), and attenuates neuroinflammation and plaque pathology in AD models (26-28). Our previous study showed that chronic administration of a low oral dose of Curc (160 ppm with food or 800 μg / day) was able to reduce inflammatory cytokine 1L-1β, carbonyl groups as oxidized protein indices, soluble Aβ and insoluble, and the plates on the APPsw Tg2576 mouse (26). Recently, Backsai et al. have discovered, using in vivo double photon microscopy to follow individual plaques in APP / PS1 transgenic mice treated with Curc, that Curc bound pre-existing deposits and reduced the average plate size by 30%, after 8 days of treatment (29).

55 Due to the proposed pharmacological actions with respect to the Curc, which include various anti-cancer, antioxidant and anti-inflammatory activities (30-33), clinical trials are underway to investigate its efficacy for the prevention of cancer and AD. Unlike α-tocopherol (vitamin E), which is a limited oxidation neutralizer associated with NO (34), Curc exerts a potent antioxidant activity for the generation of radicals associated with NO (35). Unlike NSAIDs, whose adverse side effects include ulceration and gastrointestinal bleeding, as well as liver or kidney toxicity (26, 38), Curc appears to be relatively safe, even in clinical trials on the prevention of ulcers. Curc is a biphenolic compound with hydroxyl groups in the ortho position of aromatic rings. It is separated by a β-diketone bridge, which contains two double bonds (dienone). The dienone fraction contributes to its instability in aqueous solution (39, 40), while the reduction of double bonds in vivo results in the

65 TC metabolite, which is resistant to hydrolysis (39,40). It is believed that CT reduction occurs first in a cytosolic compartment (intestinal or hepatic) (41).

Little is known about the differential effects of the Curc with respect to CT. As in the case of the Curc, it has been documented that CT exerts similar antioxidant and anti-inflammatory activities in vitro (42.43) and in vivo (kidney) (44), suggesting that CT may contribute to the benefits of the Curc. CT can be administered orally and is more easily absorbable, bioavailable and more stable than Curc (42, 44). However, CT lacks the dienone fraction, which can be subjected to the addition of Michael and may be essential for some of the effects of Curc (45, 46). It is not clear whether the CT derived from the metabolism of the Curc can have any effect, since the glucuronides of the CT, but not the free CT, are observed in plasma after the oral administration of Curc (44, 47, 48). This suggests that most plasma CT may exist in the form of glucuronide that will be quickly filtered by the kidneys.

There is also controversy regarding the bioabsorption of the Curc. A recent study found apparently insignificant levels (in nM) of Curc in human plasma, even after oral administration of a high dose (10 grams) (49). However, there are numerous reports of functional effects following oral administration of Curc. Curc pharmacokinetic studies demonstrate an avid glucuronidation and intestinal sulfation, generating uncertainty as to whether Curc or its metabolites are responsible for their beneficial actions beyond the intestine (23, 33, 47, 50-52). It has been postulated that certain conditions, such as fasting, the route of administration, the use of vehicles or the combined treatment with pepper extract (piperazine) can improve bioavailability and efficacy, although there is no consensus regarding the dosage, the frequency and route of administration, or administration together with food. To begin to clarify the interaction between bioavailability and bioefficacy, we first investigate the levels of Curc and CT in plasma and in the brain after using different routes of administration (by fasting or without fasting, ip and im), as well as its impact on Neuroinflammatory responses to an acute inflammatory stimulus (LPS). Subsequently, we investigated the relative impact of chronic administration of Curc or CT with food on Aβ and the inflammatory parameters of the AD model in the elderly APPsw transgenic mouse, as well as the presence of free CT as a product of the biotransformation of the non-glucuronidated Curc in the plasma of the mice fed with Curc.

EXPERIMENTAL PROCEDURES

Animals and treatments - The use of animals was approved by the Greater Los Angeles Veterans Administration Institutional Animal Care and Use Committee. In the acute administration paradigm, male and female C57 / B16 / J mice (3-4 months of age, n = 4) received Curc, CT or vehicle by probe (gav), i.p. or i.m., after fasting for two consecutive days, as shown in Fig. 1A. One hour after the second day dose, LPS was administered to the mice (0.5 μg / g body weight) via i.p. and they were sacrificed 3 hours later. Each of the gav and i.p. contained 148 μg (0.4 μmoles) of Curc (Cayman Chemicals, Michigan, MI) or CT (Sabinsa Corp. Piscataway, NJ) and the i.m. dose was 74 μg (0.2 μmoles). The stock solution was produced by dissolving Curc or TC in a 0.5 N solution of sodium hydroxide (NaOH), which was rapidly diluted 1: 100 and neutralized in the phosphate buffer solution (pH 7.4) described above (23). The paradigm of chronic oral administration of CT or Curc (500 mg / kg with food or approximately 1.25 mg / kg body weight) was evaluated using male and female APPsw Tg2576 mice (Fig. 1B). The mice were fed these diets for 4 months with CT (n = 5-7) or Curc (n = 5-11) during the period of accelerated plaque deposition (12-16 months of age). For the sacrifice, the mice were first anesthetized with phenobarbital (40 mg / kg body weight, i.p.) to collect a blood sample. Blood was drawn from the abdominal aorta in tubes containing EDTA and centrifuged at 1,600 x g for 20 minutes at 4˚C. The collected plasma was distributed in aliquots and subjected to instant freezing, using liquid nitrogen. Subsequently, the mice received an infusion of PBS (pH 7.4) containing protease inhibitors (20 μg / mL of pepstatin A, aprotinin, phosphoramidone, and leupeptin, 0.5 mM PMSF, and 1 mM EGTA). Brains were removed, dissected and subsequently subjected to instant freezing using liquid nitrogen and stored at -80 ° C for further analysis. In vitro studies Treatment of neurons with Aβ oligomer - The Aβ42 oligomer (100 nM) was prepared as described previously (27). In summary, the synthetic Aβ (1-42) peptide subjected to double lyophilization was monomerized by solubilization in (1,1,1,3,3,3-hexafluoro-2-propanol) (HFIP). The formation of fibrils was minimized during the subsequent evaporation of HFIP, adding sterilized water (to achieve the stock solution of 315 μg / ml). The fibrils were formed in the solution by continuous stirring (600 rpm) for 48 hours at room temperature (25 ° C). 1 mM NaHCO3 buffer solution (pH 10-11) was added to optimize the stability of the high molecular weight Aβ oligomer (HMW). The oligomers were added to the cells for 48 hours at 37 ° C with Curc (2.5 µM), TC (2.5 µM), Curc + TC (1.25 µM of each), or the vehicle. After 48 hours, LDH assays were performed on culture media, using a CytoTox 96 non-radioactive cytotoxicity assay (Promega, Madison, Wl), and the absorption at 490 nm was measured. Each data collection point was determined in triplicate and the value of S.D. It did not exceed 5%.

To assess the impact of the Curc and / or CT on the toxicity of the Aβ oligomer, the Aβ oligomers were added to SH-SY5Y human neuroblastoma cells cultured and maintained as described previously (53). The cells were placed in 96-well plates, at 10,000 cells / well, and differentiated in Dulbecco's modified Eagle's medium (DMEM) with low serum content, with N2 supplement and 10 µM trans-retinoic acid for 7 days . The medium was removed and replaced by a clean maintenance medium containing 0.1% BSA.

In vitro studies Microglia treatment with LPS - Given that LPS can induce iNOS in the CNS in microglia (54), and recently it has been documented that they induce iNOS in neurons, we subsequently investigated the effect of CT and / or Curc on iNOS in vitro. Cultures of cortical neurons from Spague Dawley rat fetuses of 18 days gestation (E18) were prepared, following the method described in Brinton et al. (56), obtaining approximately 97% of neurons and 3% of glias. In summary, the cells were first cultured in B27 medium (containing cytosine arabinonide to minimize proliferation of glial cells), supplemented with 25 μM glutamate, 0.5 μM glutamine, 5 U / ml penicillin and 5 mg / ml of streptomycin, and 10% fetal bovine serum (FBS). Three hours before treatment, the neurons were incubated in neurobasal medium (BNBM), but without B27 and glutamate. Subsequently, the cells were pre-incubated with 2.5 μM of Curc, TC, Curc + TC, or vehicle for one hour, before adding the LPS (1 μg / ml or 100 nM). The cells were collected 24 hours later. In summary, the cells were washed with cold PBS, scraped over 1 ml of PBS and then centrifuged at

5,000 rpm for 5 min. The cell granules were ultrasound in lysis buffer solution (SDS, deoxycholate, triton X100, NaCl, NaH2PO4, 0.5%, 0.5%, 1%, 150 mM, 10 mM, respectively) with a mixture phosphatase and protease inhibitor (fenvalerate, cantaridine, sodium vanadate, sodium pyrophosphate, sodium fluoride, phosphoramidon, EGTA, EDTA, leupeptin, aprotinin, pepstatin A, phenylmethylsulfonylfluoride (PMSF), 4- (2-aminoethyl sulfonyl) hydrochloride fluoride, at 0.05, 0.05, 1, 1, 50, 0.02, 1, 0.1, 0.02, 0.0015, 0.0145, 0.250, 0.5 mM, respectively). After incubation at 4 ° C for 30 min, the samples were centrifuged again (12,000 rpm for 10 min), and the supernatants were collected and used to determine the protein concentration (Bio-Rad DC protein assay kit, Hercules, CA) and iNOS levels by Western blot immunoassay. Since microglia are the main source of iNOS in the brain, we also evaluate the impact of Curc and CT on iNOS in a BV-2 microglial cell line (courtesy of Prof. Jean deVellis, UCLA) (57). BV-2 cells were grown in Dulbecco's modified Eagle culture medium (DMEM; Sigma-Aldrich, St Louis, MO, USA), supplemented with 10% heat-inactivated FBS (Gibco, Invitrogen Corporation, Carlsbad , CA) at 37 ° C in 5% CO2 in a humidified cell incubator. One day before treatment, the serum was removed and the cells were incubated in 0.1% BSA and DMEM. Subsequently, the cells were pretreated with Curc, TC, Curc + TC, or vehicle, one hour before LPS, and the cells were collected 24 hours later, as described above for cortical neuron cultures. Preparation of tissue and plasma samples for HPLC - Plasma and mouse brain samples were extracted in a 95% solution of ethyl acetate / 5% methanol and analyzed by HLPC to determine the presence of Curc (58) and TC (59), using 17β estradiol as an internal standard, as described above, except for the following modification. For the preparation of the brain tissue sample, the whole brain tissue was first homogenized in 10 vol of 1 M ammonium acetate (pH 4.6). The extraction efficiency of the Curc and CT was 88-97% and 82-90% in plasma and brain tissue, respectively. This HPLC method offered a detection limit of 10 ng / ml.

Quantitative PCR (QPCR) - iNOS mRNA brain levels were measured by reverse transcription followed by quantitative PCR (QPCR). In summary, total RNA was isolated from the brain using the RNAqueous kit (Ambion), according to the manufacturer's instructions, and treated with DNase. Next, the RNA (1 μg) was subjected to reverse transcription with dT primers, using the Retroscript kit, and divided into aliquots for individual uses. The analysis of mRNA levels was performed using a QPCR with SYBR Green PCR Master Mix (Applied Biosystems) in an SDS7700 (Applied Biosystems) thermal cycler, followed by an analysis of the dissociation curve, which verified the amplification of a single species of PCR The thermal cycler parameters were the standards for the SDS7700 (Applied Biosystems) instruments: 95 ° C for 10 min. and 40 cycles of 95 ° C for 15 sec. and 60 ° C for 1 min. Each sample was tested in triplicate. A standard curve was drawn to make relative comparisons of the sample values. The PCR primers for the iNOS were TGAGACAGGAAAGTCGGAAG (direct) and TCCCATGTTGCGTTGGAA (reverse) and normalized to the GADPH mRNA levels (a control of an internal maintenance gene) for each cDNA sample, using the reagent primers TaqMan Rodent GADPH Control Reagents (Applied Biosystems).

ELlSA assays for 8-iso-PGF2α (F2-isoprostanes), Aβ and interleukin1β (1L-1β) - For ELISAs of Aβ and IL-1β, an aliquot of frozen brain tissue (sprayed on dry ice with a mortar and a hand) was homogenized in 10 vol of TBS containing the protease inhibitor mixture (minus detergents). The samples were briefly sonicated and centrifuged at 100,000 x g for 20 min at 4 ° C. The supernatant of soluble TBS was used as a "TBS fraction" for the ELISA and was used to measure interleukin 1β and soluble Aβ. The granule was dissolved in guanidine and used to measure insoluble Aβ, as described previously (60). For the ELISA of F2-isoprostane, another aliquot of powdered tissue was used and all of the lipids were extracted using a modification of the method of Hara and Radin (8). The samples were then measured using the 8EPI-F2 Immunoassay Kit immunoassay kit (EA84; Oxford Biomedical Research, Oxford, MI), as described previously (28). Western blot immunoassay to detect iNOS, p-JNK, nitrotyrosine and carbonyl in brain tissue - To detect iNOS, nitrotyrosine and p-JNK, the TBS fraction was used. The samples (20 μg) were electroblotted on a 10% Tris-glycine gel and subsequently the gel was transferred to a PVDF membrane and blocked for 1 hour at room temperature in 5% milk with 0.05 Tween20 % in PBS. The membrane was incubated with primary mouse monoclonal antibodies at a dilution of 1: 1,000 at 4 ° C overnight: anti-iNOS antibody

(Transduction laboratories, BD Biosciences, Bedford, MA), anti-nitrotyrosine (Cayman Chemical, Michigan, MI) or antipJNK (Cell signaling; Beverly, MA). The secondary antibody was anti-mouse conjugated to horseradish peroxidase (1: 30,000), incubated for 1 hour at room temperature and developed with ECL. For carbonyl measurements, we use the "Oxyblot" kit (Intergen, Purchase, NY), as described previously (26). The TBS insoluble granule was ultrasound in lysis buffer solution (see above) with protease inhibitor mixtures and subsequently centrifuged at 100,000 x g for 20 min. At 4 ° C. The resulting supernatant was collected as a "lysis fraction". 10 μg of this fraction protein was reacted with 1 x dinitrophenylhydrazine (DNPH) for 15-30 min. and were neutralized with a solution containing glycerol and β-mercaptoethanol. These samples were electrophoresed on a 10% Tris-glycine gel, transferred and blocked. The blot was incubated overnight with an anti-DNPH mouse antibody (1: 150) at 4 ° C and subsequently incubated with goat anti-mouse antibody (1: 300) for one hour at room temperature. Bands were visualized using chemiluminescent techniques with unsaturated exposures and quantified. The resulting bands were scanned and quantified using a densiometer and Molecular Analyst software (Model GS-700, Bio-Rad Laboratories, Hercules, CA).

Immunostaining and image analysis - The 8 μM hemicerebral cryosections of Tg2576 mice subjected to a Curc and CT diet were immunohistochemically tested for Aβ (rabbit polyclonal antibody; 1: 500 anti-Aβ1-13, DAE) or GFAP (1 : 10000, monoclonal antibody, Sigma, St Louis, MO), as described above (60). In summary, antigen recovery was performed by vaporizing sections in a unmasking solution (Vector Laboratories, Burlingame, CA). After inhibiting fluorescence with 0.6% hydrogen peroxide and blocking with normal serum, the sections were incubated with primary antibodies overnight at 4 ° C. After incubating with the ABC reagent containing peroxidase (Vector Lab), the sections were developed with diaminobenzidine with enhancer metal (DAB, Pierce, Rockfoid, IL). DAE and GFAP image analyzes were performed in two coronary sections per brain. All images were obtained with a Nikon Eclipse E800 microscope (Tokyo, Japan), with a Sony Triniton video system. The images obtained were analyzed using the NIH Image public domain software to calculate the number of plates, the plate load and the percentage of the area of stained GFAP in the layers of the cortex and the hippocampus. Disaggregation of the Aβ42 oligomer by Dot-blot immunoassay - We examine the impact of CT and Curc on the disaggregation of Aβ42 in a cell-free system, using a Dot-blot with a specific oligomer antibody, as described above. (27), except for the fact that the oligomers were preformed. The freshly prepared Aβ42 oligomers (11 μM, Dr. Charles C. Glabe, UC Irvine) were matched with Curc (16 μM), TC (16 μM), Curc + TC (Curc and TC together, 8 μM each), or vehicle, for 72 hours at 25 ° C and, subsequently, at 500 ng per well, a nitrocellulose membrane was applied in a Bio-Dot (Bio-Rad) apparatus. The membrane was blocked with 10% skim milk in TBS-T at room temperature for one hour, washed with TBS-1 and probed with the anti-Aβ (6E10) or anti-oligomer (61) A11 antibody.

(1: 10,000) in 3% BSA-TBS-T overnight at 4 ° C. After washing, it was probed with a solution of anti-mouse antibody conjugated to horseradish peroxidase (Pierce) for 1 hour at room temperature. The blot was developed with SuperSignal (Pierce) for 2-5 min. The spots were scanned and analyzed with a densitometer of the GS-700 model using Molecular Analyst (Bio-Rad) software. Statistical analysis - The comparative effects of treatment with Curc and CT by HPLC were analyzed by an ANOVA analysis of a factor, followed by a post-hoc Bonfertoni / Dunn test. Western blot densiometry was analyzed by ANOVA analysis of a factor (Graph Pad InStat v.3.05, San Diego, CA). The correlation coefficient and the significance of the degree of linear relationship between parameters was determined with a simple regression model, using StatView (SAS Institute Inc., Cary, NC). All experiments were performed in triplicate or quadruplicate.

RESULTS

Previous measurements of Curc or CT using our published HPLC methods revealed a detection of CT or Curc in tissue samples containing these substances. However, it was unknown if these methods were sensitive enough to detect CT or Curc in animals treated with these substances before being slaughtered. Therefore, we provide Curc or CT using three routes of administration and using a method shown to improve its bioavailability (23). The structures of Curc and TC are shown in Fig. 2, A and B, which illustrates that Curc double bonds flanking the β-diketone bridge are lost after reduction to CT. The data demonstrated the satisfactory detection of these compounds in the treated mice. Representative chromatograms of the mice injected with Curc or CT by i.m. revealed the presence of their respective maximum values, both in plasma (Fig. 2, C and D) and in the fractions of the brain (Fig. 2, E and F), while in the control samples only the internal standard maximum value was present (Fig. 2, G and H). A detailed analysis, shown in Table 1, revealed that after acute probe administration, Curc and CT could be detected in the brain, but not in plasma. Even when the gastrointestinal barrier was avoided by injection i.p. or i.m., the plasma levels of CT were always higher (> 2 μM) than those of Curc (<0.3 μM). Even though the plasma levels of CT were markedly higher than those of Curc, the brain levels of CT were the same as those of the Curc. In fact, the plasma Curc levels of the mice treated with Curc were relatively equivalent to the respective brain levels (plasma / brain gradient close to 1), while the plasma CT levels of the mice treated with CT were several times higher than the brain (plasma / brain gradient 4.4-11). Since curcuminoid brain levels were several times lower than those used in numerous published studies, we investigated whether these levels were sufficient to exert anti-inflammatory effects. Dice

5 that the iNOS is an inflammation index that increases in the brain of the AD patient and can be inhibited by Curc at relatively low doses in vitro, we induced inflammation by injecting LPS and then analyzed the brains of mice treated with CT or Curc to detect the presence of the iNOS protein (Fig. 3, A and B) and the mRNA (Fig. 3, C). LPS caused strong increases in mRNA and iNOS protein. Although the im injection, which caused the higher brain levels of the drug, tended to be slightly more effective in attenuating responses to these LPS, all routes of administration of Curc and CT caused significant reductions of the mRNA and the iNOS protein . The regression analysis showed a close correlation between the inhibition of the iNOS mRNA and the increase in the concentrations of the curcuminoids administered, although it suggested that lower brain levels of Curc were more effective than equivalent levels of CT in reducing the iNOS mRNA (Fig. 3, D).

15 Since LPS induces CNS expression of IL-1β (2, 7), an inflammatory cytokine whose levels are known to be high in neuroinflammatory diseases, including AD, we also assess the relative impact of Curc or CT on the induction of IL-1β. Although the LPS effectively induced significant increases in brain levels of IL-1β compared to the vehicle, all routes of administration of both Curc and CT also inhibited this response (Fig. 4, A). The correlation of the levels of CT or Curc with IL-1β confirmed that CT and Curc were equally effective in reducing IL-1β (Fig. 4, B).

LPS can induce lipid peroxidation in the brain, measured by F2-isoprostanes (62), which is elevated in AD (63, 64) and in AD models (28, 65). Therefore, we investigate the impact of the Curc or TC on the

Induction of F2-isoprostanes. The results showed that LPS increased the levels of F2-isoprostanes. Although the acute oral administration of Curc or CT failed to produce sufficient brain levels to modify isoprostans, the administration i.p. slightly reduced and the administration i.m. significantly reduced levels of L2-induced F2-isoprostanes. Regression analysis of Curc levels

o CT with F2-isoprostanes shows important correlations, although lower levels of Curc appeared to cause a greater inhibition of F2-isoprostans compared to equivalent levels of CT (Fig. 4, D).

The reaction of superoxide with nitric oxide, resulting in an increase in peroxynitrite, occurs in neuroinflammatory diseases such as AD and ALS (66, 67). An indirect index of peroxynitrite can be estimated by measuring the anti-nitrotyrosine (NT) reactive proteins, which are formed by nitration mediated by

35 peroxynitrite of protein tyrosine residues. The results showed that the LPS induced a prominent NT-reactive bands of 75 kD and 90 kD, but the administration by probe of any of the compounds was not effective by modifying the NT (Fig 5, A), in the same way as the administration by probe failed to suppress induction of F2-isoprostane. On the contrary, the Curc was effective in suppressing the oxidation of LPS-induced protein, mediated by carbonyls, regardless of the route of administration (Fig. 5, B), unlike CT, which was only effective in reducing the induction of carbonyl by im injection (Fig. 5, C). Since it has been documented that fasting can enhance the absorption or use of the Curc (23, 68, 69), and since chronic treatment is probably necessary for the therapeutic intervention of diseases, we assess the impact of food ad libitum on the absorption and efficacy of Curc and CT.

45 We increase the dose administered by probe to 480 μg, which is closer to the consumption in the daily Curc diet used to reduce the pathogenesis of AD in the mouse Tg2576 (26). In this experiment, the Curc / CT ratio in the brain was 2: 3, which supports the hypothesis, unlike what happens in the plasma, that the bioavailability of the Curc in the brain is greater than that of CT Compared to fasting mice, mice previously fed ad libitum showed a significant reduction in the absorption of both Curc and CT (administered by probe), measured by their contents in the brain (Fig. 6, A). Although the Curc was not even detectable in the brain when administered with food, CT was detectable. The reduction in GI absorption derived from food intake was associated with a decrease (Curc) or non-existence (CT) of efficacy in the inhibition of cerebral iNOS induced by LPS (Fig. 6, B). However, Curc was significantly effective in the brain (p <0.05), suggesting that Curc may be effective at brain levels as limited as

55 81.3 nM (0.03 μg / 10 mg protein). Since the Curc was not detectable with acute administration by tube with food (without fasting), we subsequently investigated whether it could be detected after chronic administration in food. We select the doses indicated above to reduce oxidative damage, neuroinflammation and plaque pathology in the APPSW Tg2576 mouse (26). Table 2 indicates that, unlike acute probe administration, chronic oral administration in the Curc or CT meal resulted in detectable plasma levels of 0.095 μM and 0.73 μM, respectively. As in the case of acute administration, chronic administration resulted in plasma levels of CT several times higher than those of the Curc. Next, we evaluate the relative efficacy of the feeding with Curc and CT on the pathology of the plaques in the mouse Tg2576 APVsw. Immunostaining for anti-Aβ revealed that, compared to mice fed with

In the control, the mice fed with Curc showed a surprising reduction of the plaques (Fig. 7, A-D), similar to the one previously described (26). The fact that the diet with CT showed a slight reduction in the size of

the plaques compared to mice fed the control diet was less clear (Fig. 7, E and F). The quantification of the plate size (Fig. 7, G) and of the plate load (Fig. 7, H) confirmed that the Curc, but not the CT scan, reduced the pathology associated with Aβ. Subsequently, we analyzed the dissected cortex to detect Aβ levels and observed that animals fed with Curc, but not those fed with CT, showed a reduction of Aβ in the insoluble fraction (Fig. 7, I). The only significant impact of CT on Aβ variables occurred on soluble Aβ, which was reduced to a greater extent than with Curc (Fig. 7, J).

GFAP was analyzed immunohistochemically to assess the impact of Curc and CT on elevated gliosis in this transgenic AD model (Fig. 8, A-H). The quantification of the GFAP confirmed the known induction of GFAP in this transgenic model. Both the Curc and the CT significantly attenuated the gliosis associated with the transgene (Fig. 8). IL-1β is an index of neuroinflammation, which is believed to contribute to the pathogenesis of AD and that maintains a high level in this model. As in the case of gliosis, IL-1β was effectively reduced with both Curc and CT, compared to Tg + mice following the control diet (Fig. 9, A). However, although the carbonyls that are known to be induced by this transgene

(26) were reduced with the Curc as described above, the CT showed only an insignificant tendency to reduce carbonyls (Fig. 9, B). Since it has been documented that stress-activated (phosphorylated) cJun N-terminal kinase (pJNK) maintains high levels in cortical homogenates of this model (13) and JNK is a possible objective of the Curc, we also examine the impacts relative of the Curc and the CT on the pJNK. Western blot analysis revealed two bands at 46 and 56 kDa, which when quantified showed that both Curc and CT effectively reduced pJNK, with CT having a greater effect (Fig. 9, C and D). Since CT had more potent effects than Curc on both soluble oligomers and pJNK, the possibility of a direct relationship was assessed by a regression analysis, which showed that the degree of reduction of the soluble oligomer was directly proportional to that of reduction of pJNK (Fig 9, E). Since the data demonstrated different effects of the Curc and the CT on efficacy, and since part of the Curc can be converted into a free CT, we hypothesized that they could function synergistically in the chronic model. Therefore, the levels of Curc and CT were measured in plasma after chronic ingestion of Curc in the food. Significant amounts of CT could be detected, with Curc representing 75% and CT 25% of the total plasma curcuminoid content (Fig. 10, A). The possibility that the combination of both substances in vitro can more effectively inhibit the induction of iNOS by means of LPS was tested in neurons (55) and in a microglial cell line. The results showed that the combination of CT and Curc had a more potent effect on the suppression of iNOS than any of them separately, both on primary neurons (Fig. 10, B) and on BV2 microglia (not shown). ). We also evaluated whether this observed synergy was also applicable to the toxicity of Aβ measured by LDH. The results showed that, although both the Curc and the CT had a protective activity, their effectiveness was greater when they were administered in combination (Fig. 10, C). Although CT had a greater impact on soluble Aβ, it is known that Curc binds to amyloid and reduces its aggregation. Since there is a hypothesis that aggregation of the soluble oligomer is an important effector in AD (70), we use a cell-free assay (27) to investigate the relative impact of CT and / or Curc on oligomer disaggregation. of high molecular weight Aβ. The data showed that both CT and Curc separately were able to disaggregate the preformed oligomers, but that used together increased their disaggregation power of the Aβ oligomers (Fig. 10, D).

DEBATE

The Curc is surprisingly effective in numerous animal models in vitro and in vivo of neurodegeneration and cancer, although the blood levels necessary to produce these effects are unknown. The low bioavailability in both animals and humans has been a cause of concern, since the plasma levels necessary for many in vitro effects are not clearly achieved (41). This has generated interest in active metabolites for which Curc can act as a prodrug, such as CT, which is believed to be better absorbed and offers similar effects. However, with the exception of a few studies in vivo (44, 71), little is known about the effects of CT in vivo, and nothing about its effects on the brain. Our new conclusions demonstrate the levels in blood and target tissues necessary for various protective activities. We confirm that CT offers greater stability and absorption than Curc, although we demonstrate for the first time a lower availability of CT in the brain. We also offer evidence that supports a differential impact of CT with respect to the Curc on neuroinflammatory cascades and the pathogenesis of AD, and we provide the basis for the justification of their synergistic joint use, for example, in AD therapy. Effects of CT and Curc on GI absorption - Although Curc could not be detected in the brain after acute administration by probe of a dose of 147 μg or 480 μg, CT could be detected with the highest dose elevated administered by probe, confirming the previous suggestions that CT could have a better GI absorption (44). The improved GI absorption of the CT with respect to the Curc was also suggested by the model of chronic administration in the diet (ad libitum), where the plasma levels of CT in the mice fed with CT were almost 8 times higher than the plasma levels of Curc in mice fed Curc. Unlike what happened with acute probe administration, repeated administration on Curc's diet daily resulted in detectable plasma levels of Curc, which may be due to its ability to inhibit glucuronidation. One report showed that after a high oral dose, Curc-glucuronide and Curc-sulfate conjugates were detected, but the free Curc remained below the detection limits (50). Other studies demonstrated similar results in rats, with levels of free Curc below 5 nM (72), while our data revealed plasma levels close to 100 nM of free Curc with chronic diet administration. Pan et. to the. they found that approximately 99% of the Curc and more than 85% of the CT in the plasma are conjugated with glucuronide with acute oral administration (39). However, the Curc can inhibit

5 LTDP-glucuronosyltransferases (73). Our success in the detection of free Curc may be related to the fact that chronic administration inhibits glucuronidation or to the improvement of our detection methods (58).

It has been previously suggested that fasting was necessary to optimize the absorption of Curc after acute administration, while food could reduce the effectiveness of Curc, possibly due to the reduction of intestinal absorption (23). Our data not only confirmed that the food attenuated the absorption and efficacy of the probe-administered Curc, but also demonstrated an even greater suppression of CT efficacy. Therefore, in order to overcome the absorption reduction, a higher dose (about 2,000 μg per day) was used with the administration with the food. Increasing the rate of stomach emptying, such as that induced by solid foods, can reduce the absorption of neutral compounds, since they can be absorbed to some extent directly from the stomach (74). Since hot foods and foods with a high fat content may have an opposite effect on solid foods with regard to stomach emptying, reducing their speed (74), it is possible that in humans, unlike In mice, the Curc is better absorbed with hot foods with a high fat content, a controversy that needs to be resolved. On the other hand, another explanation of how food could reduce the absorption of Curc or CT (and its effectiveness) in mice is by stimulating gastric motility, which can decrease the absorption of some drugs, by increasing the speed of transit through the intestine (74). In summary, in addition to confirming that CT is much better absorbed in the GI tract with respect to the Curc, our data demonstrate for the first time that CT and free Curc can be detected in plasma after oral administration

25 chronic (in food) and suggest that, even when fasting is not possible, chronic administration in the diet can improve the limited oral absorption of the Curc.

Stability of the Curc and the CT in the plasma against the brain - The high ratios of the CT with respect to the Curc in the plasma show that the improved bioavailability of the CT in the plasma was not only due to an improvement in the GI absorption, since these ratios were observed in injected animals in which CT levels were between 4 (im) and 7 (ip) times higher than Curc levels, possibly due to better stability of plasma CT. This coincides with in vitro data demonstrating that CT does not degrade in PBS solution, regardless of pH, after 8 hours of incubation at 37˚C (39), while 90% of the Curc degrades before 4 hours (39, 40). The instability of the Curc as a result of hydrolysis in aqueous buffer solutions (39, 40) and

35 in plasma, regardless of the route of administration, does not necessarily imply instability in lipid compartments such as the brain, where hydrolysis can be limited. Our data suggest that Curc may be as stable or even more than CT in the brain, since Curc levels were equivalent or sometimes higher than those of CT, despite maintaining lower plasma levels. In summary, the Curc not only has limited absorption, but is expected to be highly unstable in aqueous compartments such as plasma; however, it can be more stable if administered in a lipid-rich compartment, such as the brain. Reduced absorption of CT in the brain compared to the Curc - The ratios of brain levels with respect to plasma levels reached an average of 1: 7 in the case of CT and practically 1: 1 in the case of Curc. This was surprising and may reflect differences in transport by the blood-brain barrier, the clearance

45 of the brain or plasma, resulting from differences in chemical structure, sulfation or glucuronidation. It has been noted that the total Curc (including sulfated and glucuronidated) was relatively low in the brain compared to plasma, and it has been suggested that the main factor that limited Curc's entry into the brain was its glucuronidation (75). Probably the entry of CT into the brain is also limited by glucuronidation. Our measurements of CT and free Curc in the brain compared to plasma suggest that Curc is better absorbed by the brain than CT.

Effects of CT and Curc on IL-1β and / or gliosis - Curc and CT appeared to be similarly potent inhibitors of inflammatory cytokine IL-1β and the astrocytic marker GFAP. Both Curc and CT, by all routes of administration, effectively attenuated acute induction by LPS of IL-1β at concentrations

Very limited in the brain (0.042-0.027 μg / 10 mg protein, approx. 0.1 μM), even when plasma levels were kept below the detection limit.

The regression of the inhibition of IL-1β and the Curc levels showed a precise overlap with respect to the levels of IL-1β and CT, which shows that no clear differences were observed in the potencies of these compounds. Similar to what happened in the acute inflammation model, in the chronic Tg2576 APP mouse model of brain inflammation and AD, both CT and Curc reduced IL-1β levels equivalently, which It shows that both strongly inhibit inflammation and that they achieve it at reduced levels. In multiple systems, including our previous report on the Tg2576 model (26), Curc is a potent inhibitor of inflammatory cytokine induction, including IL-1β, as reviewed in (76). This is the first demonstration of the effect of CT on lL-1β in this model. Although there is less data on the anti-inflammatory activity of CT, it is believed that the activity of the Curc is

mediated by the limitation of the activity of transcription factors AP-1 (45), NF-KB (46), or co-activators of AP-1, histone acetyltransferase p300 (HAT, (77, 78)). The inhibition of the AP-1 and NF-KB Curc causes the inhibition of cycloxygenase-2 (COX-2) in a microglial cell line (79). What is not so clear is how the Curc differs from the CT in these activities. One report suggested that compared to CT, the Curc was a

5 more potent inhibitor of COX-2 in intestinal cells (50), although another considered that its potency was similar in a macrophage cell line (80). In summary, submicromolar concentrations of CT and Curc in vivo significantly inhibit IL-1β in acute and chronic models, without a clear difference in efficacy, demonstrating that the double bonds C = C of the dienone of the Curc are not necessary for the inhibition of IL-1β in vivo. Effects of CT and Curc on iNOS expression - Similar to the great power demonstrated by CT and Curc on IL-1β and gliosis, both remarkably inhibited mRNA and iNOS protein at very cerebral levels reduced (0.1 μM). Both also significantly inhibited the induction of iNOS in microglia and neurons. Since both the Curc and the CT demonstrated so much power in vivo and at such limited concentrations, it was difficult to distinguish if there was any difference in terms of power. The first

15 The suggestion that Curc was slightly more effective than CT in the inhibition of iNOS was derived from the correlation between levels of CT versus Curc and the mRNA of iNOS, which demonstrate that above a maximum inhibition of 20 % higher brain levels of CT than Curc were required to achieve the same percentage of iNOS inhibition. Stronger evidence was observed, supporting a more potent effect of the Curc on the inhibition of iNOS, in animals without fasting in which Curc brain levels were below the detection limit, despite the administration It was effective in reducing the induction of the iNOS protein. This observation was consistent with an earlier report that revealed that in vitro Curc was more potent than CT by inhibiting iNOS in a macrophage cell line (43). A possible explanation of this more potent inhibition of iNOS by the Curc may be associated with the greater potency of the Curc by inhibiting LOX (80). IL-1β (induced by LPS or transgene) induces phospholipase 2 which can

25 trigger the activation of the lipoxygenase system, which can then trigger the expression of iNOS mediated by NF-KB (81). The anti-inflammatory activity of the Curc could also be achieved through the product of its hydrolytic decomposition, ferulic acid, but the charged ferulic acid would achieve poor penetration in the brain and would not be a more stable CT product.

CT also lacks the functionality of the α, β-unsaturated ketone (dienone), necessary to undergo Michael's addition (the nucleophilic addition of carbanions to α, β-unsaturated carbonyl compounds). The dienone fraction may be important for some of the Curc's anti-inflammatory activities, as well as for its ability to chelate metals. However, our data demonstrate that dienone is not necessary for the inhibition of iNOS, because it is not present on CT. It has been suggested that two of the main

35 anti-inflammatory activities of the Curc, the inhibition of AP-1 or NF-KB, are present in analogs without the dienone fraction, suggesting that Michael's acceptor reagent is not necessary for these activities (45, 46). Therefore, we cannot exclude an effect of AP-1 or NF-KB inhibition on the mediation of iNOS inhibition by CT. However, it seems more likely that the CT will act on the activation of the JNK upstream of AP-1 as discussed below. Antioxidant effects of CT and Curc - Although both Curc and CT effectively reduced NT formation, lipid peroxidation and protein oxidation, Curc was more effective than CT by inhibiting oxidative damage in the brain, especially carbonyls. The Curc inhibited protein oxidation at very limited brain levels (0.042 μg / 10 mg protein, about 114 nM), and inhibited both lipid peroxidation (measured by F2-isoprostanes) and NO oxidation (measured by NT) at equal brain levels or

45 greater than 200 nM. For example, brain levels of Curc after injection i.p. of Curc were the same as the CT levels after the i.p. of CT, but only Curc inhibited NT, suggesting that it may be the most potent inhibitor of NO oxidation. This activity is consistent with the observations that Curc reduced the formation of peroxynitrite in vitro (82) and in vivo (injected) in a rat model of cerebral ischemia (83). Curc in food was also effective in inhibiting lipid peroxidation in an amyloid infusion model

(28) and carbonyls in both a Tg2576 model of AD (26) and a brain trauma model (84). This report is the first to demonstrate the effectiveness of the CT on oxidative damage of the CNS. Our results that demonstrate that Curc is more potent in inhibiting lipid peroxidation than CT is surprising, given that CT acts moderately better than in vitro Curph in phantoms of membrane erythrocytes treated with tertbutylhydroperoxide (42, 87) and in alive, including enzymes from

Phase II renal detoxification nicotinamide adenine dinucleotide phosphate oxidase (NADPH) -quinone reductase and glutanione S-transferase (GST, (44)). The reasons for the discrepancies could be that the limited investigations comparing these two compounds do not take into account the role of metal chelating, which CT lacks. Chelation of metals can be critical for the reduction of oxidative damage or the pathogenesis of plaques (85). The loss of the keto-enol balance and the carbanion load that occurs with the conversion of the Curc to CT should reduce the ability of the CT to chelate metals. Studies of the structure / function of antioxidant activities suggest that 4-hydroxy-3-methoxy phenolic groups, which are common to CT and Curc, are critical for antioxidant effects (86). Another explanation of the discrepancies with the literature on the documented differential powers of the Curc and the CT on lipid peroxidation is that previous studies used in vivo doses 170 times higher than in our experiment (44), and in vitro the differential effects be

65 observed only at high doses (150 μM), which cannot be achieved by oral administration (42). The cerebral CT levels were only 1/7 of the plasma levels reducing their advantage of systemic administration. In summary, our data showed that although both Curc and CT could attenuate the induction of F2-isoprostans if brain levels are close to 0.2 μM or higher, but Curc appeared to be slightly more potent.

5 Effects of CT and Curc on Aβ in the mouse Tg2576 - Curc was more effective than CT by attenuating the pathogenesis of plaques, as suggested by the reduction in plaque size, plaque loading and reduction of Aβ levels in the detergent insoluble compartment. This was a significant conclusion, which demonstrates a clear and important difference between the two compounds. The inhibitory effect of the Curc on the formation and disaggregation of fibrils and oligomers had previously been demonstrated in a different animal cohort (27) and the potent effects against the formation or disaggregation of fibrils had been observed previously (88-90) ; However, this is the first time that a differential effect of Curc versus CT in vivo has been documented. It is interesting to note that it has been documented in vitro that, although the addition or elimination of the methoxy groups of the phenolic rings did not attenuate the potency of the curcuminoids to reduce the formation of fibrils, the loss of the dienone double bonds resulted in a loss of inhibition of the fibrils

15 (88). Taken together, these data raise the possibility that the dienone bridge is not only critical for the inhibition of amyloid aggregation in vivo, but is also necessary for a maximum reduction of amyloid plaques in vivo. Curc is less stable than CT and, therefore, it can be metabolized more rapidly in ferulic acid, which also affects the formation of fibrils; However, EC50 concentrations for the disaggregation of fibrillary Aβ are about 10 times higher for ferulic acid than for Curc (91), which contradicts the existence of a substantial role of the metabolite in terms of the greater efficacy of Curc on the plates and insoluble Aβ in vivo. However, factors other than amyloid binding may explain the more potent effects of Curc in terms of reducing plaques in vivo, such as metal chelating, which is known to reduce plaque formation (85 ). Although CT does not bind well to metals (see above), Curc easily binds to Cu2 + and Fe2 + (92). The spectrophotometric measurements of the

The union of the Curc to metals demonstrates an affinity in a range comparable to that of Aβ by metals, suggesting that Curc could compete for metals or eliminate Aβ metals (92). In summary, the Curc, although not the CT, reduced insoluble Aβ and plaque loading, possibly due to the Curc's ability to bind directly and disaggregate the fibrils, and / or by the ability of the dienone bridge to chelate metals This supports the use of Curc, to the detriment of CT, for the therapeutic reduction of plaque deposition of AD.

Our data on soluble Aβ levels confirmed our previous report on the effects of Curc in reducing soluble Aβ on Tg2576, showing a similar decrease of 50% (26). Our soluble Aβ levels (TBS) in the elderly Tg2576 mice are well above baseline production levels of the Aβ monomer in the Tg2576 and clearly include soluble aggregates (oligomers) such as those shown in Western blot immunoassays. . It could be presumed that, if insoluble and soluble Aβ compartments were in equilibrium, drugs that reduce Aβ aggregation or increase Aβ clearance would similarly reduce both amyloid fibrillar plaques and soluble oligomers, as we observed in the case of the Curc. The surprising conclusion was that CT did not have any statistically significant effect on plaques or insoluble Aβ, although it reduced soluble Aβ by 75%, more effectively than Curc. One possibility is that, despite having a very similar structure and a greater affinity of the Curc for the fibrils (88), the CT may have a higher binding affinity for the oligomers. However, in the cell-free assay, Curc and CT were equally potent by disaggregating oligomers of high molecular weight. A second possibility is that CT serves as a peripheral sink for CNS oligomers, similar to the brain's Aβ clearance after peripheral administration of anti-Aβ antibodies (93, 94). Unlike the

45 Curc, where the plasma / brain gradient was close to 1, in the case of CT, the plasma / brain ratios ranged between 4: 4 (i.m.) and 11: 1 (i.p.); therefore, CT may be able to sequester the Aβ oligomers and favor their exit from the CNS. A third possibility is that the inhibition of JNK by CT reduces the production of Aβ, since recent data have shown a reduction in Aβ production with inhibitors of JNK in vitro (95) and with the JNK knockout mouse crossed with the APP Tg mouse in vivo (96). In summary, the reduction of soluble Aβ by the Curc, and particularly of CT, is of great relevance for intervention in AD, since the plaques do not correlate well with cognitive impairment (97) and high oligomers Molecular weight are involved in synaptic loss and neurodegeneration in AD (61, 98-102).

Effects of CT and Curc on pJNK in the mouse Tg2576 - Curc seems to address inflammatory cascades at

55 through multiple mechanisms, by inhibiting: 1) NF-KB blocking the degradation of 1kBα and the nuclear import of NF-KB, 2) AP-1 directly blocking its interaction with its DNA binding motif (103), and 3) AP-1 inhibiting upstream mediators, such as the activation of JNK, at lower doses and easier to achieve (approx. 10 nM or less) (20). The Curc inhibits the JNK through an unidentified upstream mechanism, since the Curc cannot directly interact with the JNK or direct upstream mediators SEK1, MEKK1 or HPK1 (104). We observed that the CT scan was surprisingly better than the Curc (p <0.002) inhibiting JNK in vivo, in the Tg2576 model. This is the first report that CT can inhibit JNK, which clearly demonstrates that the addition of Michael is not necessary for the inhibition of JNK in vivo. Although soluble Aβ, which experienced a markedly greater reduction with CT (not fibrillary Aβ), was very closely correlated with JNK, it is unclear whether the JNK reductions associated with

65 curcuminoids contributed to the reduction of oligomers. It is hypothesized that the activation of JNK, a tau kinase, mediated in the final phase of neurodegeneration (105), is regulated upwards in AD (11) and models

transgenic AD (12, 13) and may mediate the inhibition of the Aβ oligomer of LTP without loss of neurons (106). JNK can be easily activated by reactive oxygen species (ROS) that cause oxidative damage, at least through 4 different cystosolic systems, as well as by the membrane lipid raft system (10). In summary, Curc can not only inhibit JNK in vitro (20), but also in vivo. Our data demonstrate for the first time that CT can also act as a JNK inhibitor in vivo, and with more power than Curc. Both are closely correlated with reductions in soluble Aβ, although it is still unclear whether the CT acted directly on the JNK system, as demonstrated in the case of the Curc in many other systems, or if it acts through the reduction of soluble Aβ, or both. CT / Curc synergies - The free CT detected in the plasma of the mice fed with Curc was approximately 25% of the free Curc levels. It has been documented that Curc is reduced in vitro to free CT. In the cytosol of intestinal tissue incubated with Curc (41), there are no previous reports of free CT in vivo after treatment with Curc. In vivo, CT glucuronides or total CT after β-glucuronidase have been detected in the plasma of mice injected (ip) with Curc or in the serum (44) or plasma (48) of mice fed with Curc. However, curcuminoids-glucuronides are eliminated very quickly by the kidney and may not have any bioefficacy. An explanation of our ability to measure the most relevant free CT from a functional point of view in animals treated with Curc may be the increased sensitivity of our published trial (59). In summary, our observation of free CT in mice fed with Curc is the first demonstration of Curc bioconversion in free CT, and suggests its presence at a concentration high enough to contribute to efficacy. The observations that Curc was better reducing the pathogenesis of plaques and carbonyls, while CT was better reducing soluble Aβ and pJNK, raised the question of whether Curc and CT could contribute to efficacy and if they could really act synergistically in chronic models. We demonstrate an improved synergistic effect of CT and Curc when co-administered with respect to the administration of each of them separately (at the same total concentration of curcuminoid) in three systems: 1) attenuation of the LN-induced iNOS in primary neurons and a microgliagle cell line; 2) blocking of Aβ toxicity in a neuroblastoma model; and 3) disaggregation of preaggregated high molecular weight oligomers. Conclusions - Our data show that, despite the gastrointestinal absorption and / or markedly superior stability of the CT scan, it obtained plasma levels between 4.3 (im) and 6.7 (ip) times higher than equal doses of Curc , the brain levels obtained with the Curc were similar or even higher than those of the CT. Both the Curc and the CT, at nanomolar levels, attenuated the acute induction by LPS of the iNOS, NT, carbonyls and lipid peroxidation. This occurred despite evidence showing that, unlike the Curc, CT does not inhibit possible anti-inflammatory targets through the addition of Michael. Thus, in the chronic amyloid model of AD, CT and Curc had similar effects on the reduction of IL-1β and GFAP. It is interesting to note that CT and Curc had different effects on Aβ, since only Curc reduced the pathogenesis of plaques and CT had a more potent effect on reducing the activation of soluble Aβ and JNK. Finally, the CT and the Curc appeared to exert differentiated and synergistic effects on the pathogenic systems of AD, which supports the justification for the joint use of the CT and the Curc as an intervention strategy in the AD.

*FOOTNOTES

AG028583 (SAF), AG021975 (SAF), AG016570 (Cummings. Project 1. GMC). We appreciate the original support of Mrs. K.K. Siegel with respect to curcumin, 10 years ago, at a UCLA Pilot Center beneficiary of a grant for research related to aging.1 The abbreviations used are the following: Curc, curcumin; TC, tetrahydrocurcumin; i.p., intraperitoneal, i.m., intramuscular, NFKB, nuclear factor kappa B, Aβ, beta-amyloid peptide; iNOS, inducible nitric oxide synthase; LPS, lipopolysaccharide; JNK, cJun N-terminal kinase; NT, nitrotyrosine; GFAP, acidic fibrillar glial protein; IL-1 β interleukin-iβ; AD, Alzheimer's disease; ROS, reactive oxygen species; RNS, reactive nitrogen species; CNS, central nervous system; COX, cyclooxygenase; HAT, histone acetyltransferase; NSAIDs, non-steroidal anti-inflammatory drugs.

LEGENDS OF THE FIGURES

Fig. 1. The acute and chronic administration routes of Curc and CT in mice were used to investigate the

bioabsorption and bioefficacy. A, mice were treated with Curc or CT by probe (gav), injection

intraperitoneal (i.p.), or intramuscular injection (i.m.), using doses of 0.4, 0.4, or 0.2 μmoles (148, 148,

73.4 μg), respectively. B, Tg2576 APPsw elderly transgenic mice were fed ad libitum diets with or without Curc and CT for four months at 500 ppm in the meal (500 mg / kg meal).

Fig. 2. HPLC chromatograms demonstrated a detection of Curc or CT in plasma and in the brain after i.m. injection, but not in control tissue extracts. The structures of the Curc (A) and the CT (B) with circles that highlight the structural differences between them in the diketone bridge. The wavelengths for the UV detection of the Curc and the CT were 262 nM and 280 nM, respectively. Chromatograms of plasma HPLC (C, D) and brain (E, F) are shown for Curc and CT, respectively. The Curc was detected at a retention time of 5.56 minutes and the internal standard (ISD)

at 10,898 minutes (C E), while CT was detected at 10.6 minutes and ISD at 19.9 minutes

(D, F). Control samples only showed the maximum level of ISD, with no maximum levels of CT

nor Curc (G, H). Arrows indicate maximum retention times. The chromatograms shown are

representative of the analyzes performed with plasma or brain tissue of three or four mice. mAU

5

indicates miliunity of absorbance.

Fig. 3.

 Curc and CT suppressed the mRNA and the iNOS protein induced by LPS. Mice are given

they injected the LPS or the vehicle, they were sacrificed 4 hours after the administration of the Curc (A)

or CT (B), and the supernatant of brains extracted in TBS was subjected to Western electrophoresis

blot and immunostaining with anti-iNOS and β-actin. Representative lines and their quantification are shown

densiometric C) RNA was extracted from the brain and measured to detect iNOS mRNA by RT-PCR

quantitative D) The percentage of inhibition of the iNOS was subject to regression with respect to the

Curc and CT concentrations. Closed circles, Curc; open circles, TC. The values shown are

the amount of iNOS mRNA in the form of the mean ± SD. * p <0.05, ** p <0.01 and *** p <0.001 represent

fifteen

a significant difference compared to positive controls (LPS treatment alone; n = 4).

Fig. 4.

The acute injection of CT and Curc similarly attenuates the induction of IL-1β and F2-isoprostane,

although with probe only IL-1β is suppressed. (A) The quantification of IL-1β levels in

Mouse brain homogenates were determined by double phase ELISA. The rats

pretreated with or without Curc and CT show that the injection of LPS (i.p.) increases IL-β levels

more than 6 times, an effect that is partially suppressed (> 50%) in the acute administration of Curc

or CT, regardless of the route of administration. (B) The correlation of the Curc or TC levels

with the percentage inhibition of IL-1β it was statistically significant. (C) The levels of 8-iso-

PGF2a were measured in frozen powdered brain lipid extracts by ELISA to evaluate

25

Lipid peroxidation and was shown to be increased by 40% by injection of LPS. Despite

that acute administration by CT or Curc probe was not sufficient to affect peroxidation

lipid of the brain, induction of lipid peroxidation was partially suppressed by injection

i.p. and completely suppressed by the i.m. injection of any of the compounds. (D) The correlation

Between brain levels of Curc or CT and brain levels of F2-isoprostane was significant. Do not

However, Curc produced a greater inhibition of F2-isoprostane than CT when levels

reached rose above 0.075 μg / mg protein. The values shown are the mean ± SD. *

p <0.05, ** p <0.01 and *** p <0.001 represent a significant difference compared to controls

positive (LPS treatment alone; n = 4).

35

Fig. 5. Acute administration of Curc or CT (by i.m. injection) similarly suppresses induction by LPS

of nitrotyrosine (NT) in the brain, but the probe administered Curc or i.p. is more effective than CT

in the suppression of carbonyls. (A) NT proteins of mouse brain homogenates are

measured by Western blot immunoassay with anti-nitrotyrosine antibody. The top panel shows

a representative western blot of nitrotyrosine and β-actin control. The densiometric quantification of

bands and normalization to β-actin showed an induction more than 7 times higher than NT by

LPS injection with respect to the vehicle. Although acute administration by Curc or CT probe

showed no effect on the induction of NT, the i.m. injection of any of the compounds

suppressed the induction of NT. The intraperitoneal injection of Curc partially suppressed the induction of

NT, while the i.p. of CT did not affect the induction of NT. Oxidized protein levels

Four. Five

of the brain were determined in the supernatant extracted from the lysis of the insoluble granules in TBS,

using an Oxyblot analysis with an anti-DNP antibody. The impact of Curc (B) or CT (C) on

Carbonyl groups of the oxidized proteins are shown by representative lines on the left.

The quantified values on the right are the mean ± SD. * p <0.05 and ** P <0.01 represent a difference

significant compared to positive controls (LPS treatment alone; n = 4).

Fig. 6.

The brain and efficacy levels of the probe-administered Curc and CT are reduced if the mice

Do not fast. 480 μg of Curc or CT per probe were administered to the mice under fasting conditions

(FAS) or without fasting (nFAS) and subsequently, after 4 hours, the brains were removed and

prepared for the analysis of Curc and CT levels by (A) HPLC and (B) iNOS protein by

55

Western analysis where the upper panel showed a representative western blot of the iNOS and β

actin (endogenous control). When mice had food, Curc levels were not

detectable and the levels of CT were reduced by 50% compared to the levels obtained on an empty stomach. The

Curc and CT were not as effective in reducing iNOS when mice had food. The

Values shown are the mean ± SD. * p <0.05 and ** p <0.01 represent a significant difference in

comparison with the control (with, untreated, n = 4) Curc or CT per probe n = 4, each treatment (ND, no

detectable)

Fig. 7.

Curc administered in the diet seems more effective than CT by reducing the pathology of plaques in

the mouse Tg2576 APPsw, although both reduce soluble Aβ levels. Curc or CT was administered to

65

Elderly Tg2576 mice for 4 months (12-16 months old) with food (500 ppm)

during accelerated plate deposition. Representative micrographs stained with antibodies

for Aβ (DAE, anti Aβ 1-13) showed that, unlike Tg + mice subjected to the diet of

control (A and B), the mice with the Curc diet demonstrated a marked reduction in the number and

plate size (C and D). However, sections of mice with the CT diet (E and F)

demonstrated a distribution of plaques similar to that of mice with the control diet. In spite of

5

that the quantification of the image analysis confirmed that Curc's diet could reduce the size of

the plates (G) and the plate load (H), the CT diet seemed to have no effect on the pathology of

the plates. Biochemical analysis of guanidine-insoluble Aβ of cortical homogenates (I)

using Aβ double phase ELISA revealed similar results, demonstrating the efficacy of the

Curc and the inefficiency of CT reducing insoluble Aβ levels. On the contrary, both the Curc and the

CT suppressed soluble Aβ levels in brain homogenates, with CT being the most potent (/)

Increase bar = 75 μm. Statistical analysis of the data was performed using two-way ANOVA.

and the values shown are the mean ± SD. * p <0.05, ** p <0.01 and *** p <0.001 represent a difference

Significant control diet (polyunsaturated fatty acid reduction diet n-3; n = 5) and diet

standard (n = 5) compared to the treatment of Curc (n = 9) and CT (n = 7) of Tg + mice,

fifteen

respectively.

Fig. 8.

Curc or CT diets improved glial activation dependent on Tg2576. Micrographs

demonstrate that, unlike the brains of the Tg- (A and B) mice, the brains of the Tg + mice

(C and D) demonstrated an increase in staining for GFAP. Unlike fed Tg + mice

with the control diet, Tg + mice fed with Curc (E and F) or CT (G and H) in food

demonstrated a reduced GFAP (glial activation). Quantification of the GFAP staining percentage

demonstrated significant attenuation of transgene-dependent gliosis, both with CT and with

Curc. The values shown are the mean ± SD. ** p <0.01 represents a significant difference of

Curc treatment (n = 9) or CT (n = 7) in mice compared to the control diet (n = 5).

25

Increase bar = 75 μm.

Fig. 9.

In APPsw Tg2576 mice, both Curc and CT diet reduced IL-1β and phosphorylation

of the cJun N-terminal kinase (pJNK), while the Curc, although not the CT, reduced

carbonyls (A) Unlike Tg + mice with the control diet, Curc-fed mice

or CT demonstrated a 25% reduction in IL-1β levels measured by an ELISA assay of

double phase of the supernatant fraction extracted from the TBS of the brain homogenate. (B) Unlike

of the Tg + mice with the control diet, the presence of Curc in the brain reduced the carbonyls

measured by Western blot with anti-DNP antibody of the buffer extracts with detergent

the homogenates of the brain. Although the mice fed with CT showed a

35

reduction tendency, this was not statistically significant. (C) Unlike mice

fed with the control diet, those fed with CT or Curc demonstrated a reduced pJNK:

Representative lines show the 46 and 56 KDa bands. Densitometric quantification demonstrated

that compared to Tg + mice fed the control diet, mice fed with

Curc and CT showed reductions in pJNK, with CT showing a greater reduction. The values

shown are the mean ± SD. * p <0.05, *** p <0.01 and *** 0.001 represent a significant difference of

Curc treatment (n = 9) or CT (n = 7) in Tg + mice with the control diet (n = 5).

Fig. 10.

Bioconversion of the Curc to CT in vivo and its synergies reducing the iNOS, the toxicity of Aβ and the

aggregation of Aβ in vitro. (A) Plasma samples were collected from fed mice for 4 months

Four. Five

ad libitum with Curc in the food (500 ppm, n = 5). Plasma chromatograms on HPLC columns

C18 reverse phase show maximum Curc and CT using wavelengths of 262 and 280 nm,

respectively. The arrows show maximum retention times of the authentic Curc and TC at 5.7 and

10.9 min, respectively, and the internal standard (ISD) at R = 11.8 and 21.1 min., Respectively. (B)

Anti-inflammatory synergy of Curc and CT on iNOS. Cultures of primary cortical neurons

(B) and the BV-2 microglial cell line (not shown), were treated with LPS (50 μg / ml), and

Induced protein levels of iNOS were measured in the detergent soluble fraction with or without Curc

(2.5 μM), TC (2.5 μM), Curc + TC (1.25 μM of each), or vehicle. The top panel showed the lines

representative of the Western blot of the iNOS and β-actin (endogenous control). Quantification

densiometric of the iNOS band showed the induction of iNOS by expected LPS, which was attenuated by

55

Curc (50%) and CT (80%), but completely suppressed by the combination of Curc + TC (95%). (C)

The impact of previous exposure with Curc and / or CT on the AP42 oligomer (500 nM) -induced death

SH-SY5Y neuroblastoma cell, after 48 hours of exposure, using the LDH of the media as

toxicity index Aβ treatment increased LDH release (normalized according to

percentage of maximum LDH), which could have been attenuated with Curc (in 40%) or CT (in 20%). Without

However, the Curc together with the CT suppressed the toxicity of Aβ (by 70%), significantly reducing the

suppression achieved by any of them individually. (D) The effect of the Curc and / or CT on the

preaggregated oligomers (11 μM) in a cell free assay. The Curc, TC, Curc + TC (molar ratio of

1.45 to 1), or the vehicle was matched and agitated for 72 hours at room temperature with the

Aβ oligomer, before loading 500 ng of protein in a nitrocellulose membrane in an apparatus of

65

Dot blot. The impact on oligomerization was determined by an oligomer antibody

specific, A11, while total Aβ was measured in separate immunoassays, using the antibody

of Aβ 6E10. The upper panels show representative Dot blot immunoassays and the lower panel graphs the quantitative analysis of the optical densities (OD) of the Dot blot. The CT was better than the Curc by reducing the staining of the Aβ oligomer aggregate and together they obtained the greatest effect, apparently disaggregating the Aβ oligomer. The values shown are the means ± SD; * p <0.5, ** p <0.01, and *** p <0.001 represent significant differences compared to the positive control (n = 4) and treatments (n = 4).

References

Table 1

Curc and CT can be detected in the brain after administration by tube or injection, and in plasma

5 only after administration by injection. These data showed brain and plasma levels of Curc or CT, 4 hours after administration (by probe, i.p., or i.m.). Data are presented as mean ± SD. The different letters of the superscripts mean statistical difference of means between each other, between columns (treatments), within the rows or between rows (paths) within the treatments. p <0.05 SD; standard deviation; ND; not detectable.

TC Curc Treatment

Dose Plasma Brain Plasma Brain μg (μg / ml) (μg / 10 mg of (μg / ml) (μg / 10 mg of

protein) protein) With 0.0 ND ND ND ND Gav 147.0 ND 0.042 ± 0.019a ND 0.027 ± 0.002a

i.p. 147.0 0.127 ± 0.035d 0.074 ± 0.003b 0.847 ± 0.019a 0.076 ± 0.003b (0.345 μM) (2.302 μM)

i.m. 73.5 0.238 ± 0.048C 0.116 ± 0.009d 0.971 ± 0.092a 0.223 ± 0.023C (0.647 μM) (2.639 μM)

Table 2

After chronic administration of Curc or CT in the diet, the respective plasma levels are easily detectable, particularly in the case of CT.

15 APPsw mice were fed with Curc or CT in the food at 500 ppm for 4 months, before analyzing the presence of Curc or CT in plasma, respectively. Data are presented as mean ± SD. The different letters of the superscripts mean statistical difference of the means from each other; p <0.001. SD; standard deviation, ND; not detectable.

Treatment Dose in mg / kg of Plasma food (μg / ml)

(daily consumption mg / day) With 0.0 (0) ND Curc 500 (2.5) 0.035 ± 0.014a

(0.095μM) TC 500 (2.5) 0.270 ± 0.003b (0.734 μM)

Claims (24)

1. A curcuminoid composition that offers improved oral bioavailability, comprising: a curcuminoid; an antioxidant; a pharmaceutically acceptable, water-solubilizing vehicle, comprising:

(to)

lipid micelles where the composition is provided in the form of microemulsion, solid lipid nanoparticles or lipid micelles; or

(b)

 a microencapsulated oil where the composition is provided as a microencapsulated oil;

optionally, a glucuronidation inhibitor, and where the composition is suitable for oral administration.

2.

A curcuminoid composition as mentioned in claim 1, wherein the curcuminoid is selected from the group consisting of curcumin, tetrahydrocurcumin, demethoxycurcumin, bidemethoxycurcumin, curcumin esters and combinations thereof.

3.

A curcuminoid composition as mentioned in claim 1, wherein the curcuminoid is a curcumin ester comprising the esterification product of curcumin and a monocarboxylic acid.

Four.

 A curcuminoid composition as mentioned in claim 3, wherein the monocarboxylic acid comprises an essential fatty acid.

5.

 A curcuminoid composition as mentioned in claim 4, wherein the essential fatty acid comprises an Omega-3 fatty acid.

6.

A curcuminoid composition as mentioned in claim 5, wherein the Omega-3 fatty acid is docosahexaenoic acid.

7.

A curcuminoid composition as mentioned in claim 1, wherein the vehicle comprises lipid micelles and these lipid micelles are formed by one or more fatty acids, phospholipids, bile acids, edible oils, mixtures thereof, and / or pharmaceutically acceptable salts, hydrates or conjugates thereof.

8.

 A curcuminoid composition as mentioned in claim 7, wherein the vehicle comprises phosphatidylcholine.

9.

 A curcuminoid composition as mentioned in claim 7, wherein the vehicle comprises an edible oil selected from the group consisting of olive oil, rapeseed oil, fish oil and combinations thereof.

10.

 A curcuminoid composition as mentioned in claim 1, wherein the vehicle comprises a microencapsulated oil and the microencapsulated oil is an edible oil selected from the group consisting of olive oil, rapeseed oil, fish oil and combinations thereof.

eleven.

A curcuminoid composition as mentioned in claim 1, wherein the antioxidant is selected from the group consisting of ascorbic acid, acylated fat soluble ascorbic acid, vitamin C, alpha-lipoic acid, Nacetylcysteine, reduced glutathione, tetrahydrocurcumin and combinations thereof.

12.

A curcuminoid composition as mentioned in claim 1, wherein the antioxidant comprises ascorbic acid.

13.

 A curcuminoid composition as mentioned in claim 1, wherein the antioxidant is both water soluble and lipid soluble.

14.

 A curcuminoid composition as mentioned in claim 1, wherein the glucuronidation inhibitor is selected from the group consisting of tetrahydrocurcumin, piperine, probenecid, diclofenac and combinations thereof.

fifteen.

 A curcuminoid composition as mentioned in claim 1, comprising:

curcumin; tetrahydrocurcumin; phosphatidylcholine; Docosahexaenoic acid; Y an antioxidant, 5 where the composition is provided in the form of a plurality of lipid micelles. 16. A curcuminoid composition as mentioned in claim 1, comprising: curcumin; 10 phosphatidylcholine; Docosahexaenoic acid; and an antioxidant; where the composition is provided in the form of lipid micelles. fifteen 17. A curcuminoid composition as mentioned in claim 1, comprising: curcumin; Docosahexaenoic acid; 20 an antioxidant; where the composition is provided in the form of lipid micelles. 18. A curcuminoid composition as mentioned in claim 1, wherein the curcuminoid comprises a product of the esterification of curcumin and docosahexaenoic acid. 19. A curcuminoid composition as mentioned in claim 1 having the form of solid lipid nanoparticles comprising an internal phase lipid, a surfactant and a co-surfactant. 20. A curcuminoid composition as mentioned in claim 19, wherein the internal phase lipid is selected from the group consisting of triacylglycerols, such as tricaprine, trilaurin, trimiristine, tripalmitin, tristearin; acylglycerols, such as glycerol monostearate, glycerol behenate, glycerol palmitosterate; fatty acids, such as stearic acid, palmitic acid, decanoic acid, behenic acid; waxes, such as cetyl palmitate; cyclic complexes, such as cyclodextrin, para-acyl-calix-arenes; and combinations thereof. 21. A curcuminoid composition as mentioned in claim 20, wherein the surfactant is selected from the group consisting of phospholipids, such as soy lecithin, egg lecithin, phosphatidylcholine; propylene oxide / ethylene oxide copolymers, such as Poloxamer 188, Poloxamer 182, Poloxamer 407, Poloxamine 908; propylene oxide / sorbitan ethylene oxide copolymers, such as Polysorbate 20, Polysorbate 60, Polysorbate 80; 40 polymers of alkylaryl polyether alcohol, such as tyloxapol. 22. A curcuminoid composition as mentioned in claim 21, wherein the co-surfactant is selected from the group consisting of sodium cholate, sodium glycocholate, sodium taurocholate and sodium taurodeoxycholate. 23. A curcuminoid composition according to any one of claims 1 to 22 for use in the treatment of a neurodegenerative disease associated with age, wherein said composition is administered orally. 24. A composition as mentioned in claim 23, wherein the curcuminoid is selected from the group consisting of curcumin, tetrahydrocurcumin, demethoxycurcumin, bidemethoxycurcumin, curcumin esters and combinations thereof. 25. A composition as mentioned in claim 23, wherein the therapeutically effective dose is sufficient to achieve a blood curcuminoid concentration of 0.1-2 micromolar. 26. A curcuminoid composition according to claim 10, wherein the pharmaceutically acceptable water-soluble vehicle comprises an edible oil encapsulated for use in the formation of lipid micelles in vivo, wherein the composition is administered to a subject during use.